System and method for counting occupants of room

Abstract

To provide a system and a method for counting the occupants of a room that are able to accurately correct the respective numbers of persons entering and leaving a floor and are able to accurately count the occupants of the room on the floor.SOLUTION: A system for counting occupants of a room includes: a difference calculation unit 25 that calculates a difference between a total number of entering persons and a total number of exiting persons on a floor of a building, based on operation data of an elevator; a correction process data calculation unit 26, for time-series data of the number of entering persons and the number of exiting persons on the floor, that calculates correction process data for correcting at least one of the number of entering persons and the number of exiting persons on the floor in a predetermined time, based on the difference between the total number of the entering persons and the total number of the exiting persons; and an online calculation unit 30 that corrects the number of the entering and exiting persons on the floor for which the correction process data have been calculated, using the operation data of the elevator, and counts the occupants of the room.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a system and a method for counting the number of people in a room, and more particularly to a system and a method for counting the number of people in a room on each of a plurality of floors accessible by elevator. [Background technology]

[0002] In buildings with multiple floors, such as office buildings, diverse work styles have become prevalent on each floor, and the traditional uniform operation of equipment on each floor is no longer appropriate. In other words, it is now necessary to operate equipment on a floor-by-floor basis, rather than uniformly. To do this, it is necessary to know the number of people occupying each floor.

[0003] Patent Document 1 discloses a resident population estimation device. This resident population estimation device has a boarding / alighting history information generation unit that generates boarding / alighting history information, including the floors and boarding / alighting dates and times of each elevator user, based on the boarding and alighting status of elevator cars installed in a building. This resident population estimation device also has a correction parameter calculation unit that calculates correction parameters used to calculate the number of people occupying each floor based on the number of people entering and exiting the building within a predetermined period and the boarding / alighting history information within the predetermined period. This resident population estimation device also has a resident population estimation unit that estimates the number of people occupying each floor, including people not currently using the elevator, by correcting the number of elevator users occupying each floor, obtained from the boarding / alighting history information corresponding to elevator users who boarded and alighted the elevator car within a resident population estimation period from a reference time point to the present, using the correction parameters.

[0004] Patent Document 2 discloses an elevator dispatch planning system. This elevator dispatch planning system includes a movement prediction device that generates out-of-home status prediction data by statistically processing the power consumption patterns of each section on each floor of a residential building; a dispatch plan prediction device that creates a dispatch plan for elevators to wait at appropriate floors based on the out-of-home status prediction data; and an elevator control device that executes dispatch control based on the dispatch plan. The elevator control device collects passenger history data that identifies the floor from which passengers actually boarded the elevator after waiting at the dispatch floor through dispatch control. The dispatch plan prediction device corrects the current dispatch plan based on the passenger history data to create a new dispatch plan for elevators to wait at more appropriate floors, and executes dispatch control based on the new dispatch plan. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2016-74525 [Patent Document 2] International Publication No. 2017 / 175380 Summary of the Invention [Problem to be solved by the invention]

[0006] By using elevators in buildings as people flow sensors, it becomes possible to grasp the number of people on each floor without installing new sensors, and to efficiently operate air conditioning, lighting equipment, cleaning, security, etc. However, conventional methods do not adequately grasp errors in the number of people entering and leaving a facility. There is a particular problem of large errors in the number of people using the stairs rather than the elevator. Furthermore, it is difficult to calculate the number of people on floors where elevator operation data cannot be obtained. The present invention aims to provide a number of people counting system and method that can accurately correct errors in the number of people entering and leaving a floor and accurately calculate the number of people present on the floor. [Means for solving the problem]

[0007] In order to solve the above problems, the present invention provides a people counting system that includes a people difference calculation unit that calculates the difference between the total number of people entering and the total number of people leaving on a floor of a building based on elevator operation data, a correction processing data calculation unit that calculates correction processing data to correct at least one of the number of people entering and the number of people leaving on the floor at a specified time based on the difference from the total, and a people occupancy calculation unit that corrects the number of people entering and leaving on the floor that has calculated the correction processing data based on the elevator operation data, thereby calculating the number of people occupying the floor.In this case, it is possible to provide a people counting system that can accurately correct errors in the number of people entering and leaving on a floor and accurately calculate the number of people occupying on a floor.

[0008] Here, for example, the correction processing data calculation unit sets multiple movement models as models of people moving between floors, calculates model correction processing data as correction processing data for each movement model, and obtains correction processing data from the model correction processing data. In this case, by using a movement model that matches the actual movement of people, it is possible to calculate more accurate correction processing data. Furthermore, for example, the correction processing data calculation unit determines the ratio at which multiple movement models are applied, allocates the difference according to the determined ratio, calculates model correction processing data from the allocated difference, and sets the sum of the calculated model correction processing data as the correction processing data. In this case, a ratio corresponding to the actual movement of people on each floor can be set for the movement model, and more accurate correction processing data can be created. Furthermore, for example, the correction processing data calculation unit changes the ratio for each floor, and in this case, it is possible to set a ratio that corresponds to the actual movement of people on each floor. Furthermore, for example, when calculating correction processing data for correcting the total number of people exiting, the correction processing data calculation unit uses, as movement models, a model in which elevators are used but stairs are not used, a model in which elevators are used when going up and stairs are used when going down, a model in which stairs are used when the elevators are crowded and elevators are used otherwise, and a model in which elevators are not used but stairs are used, and changes the proportion at which these movement models are applied for each floor. In this case, the movement models can be made to correspond to the actual movement of people. For example, the correction processing data calculation unit increases the proportion of models that use stairs and not elevators for lower floors, which corresponds to the actual situation where people tend to use stairs and not elevators for lower floors. Furthermore, for example, when the total number of people leaving is smaller than the total number of people entering, the correction processing data calculation unit calculates correction processing data to correct the number of people leaving, and when the total number of people entering is smaller than the total number of people leaving, the correction processing data calculation unit calculates correction processing data to correct the number of people entering. In this case, it is possible to correct the error between the number of people entering and the number of people leaving. Furthermore, for example, for floors for which elevator operation data cannot be collected, the occupancy calculation unit calculates the number of occupants using time-series data of physical quantities related to the number of occupants on the floor. In this case, the number of occupants can be calculated even for floors for which elevator operation data cannot be collected. Furthermore, for example, the correction processing data calculation unit determines the relationship between the number of occupants and a physical quantity related to the number of occupants on a floor for which elevator operation data can be collected, and the number of occupants calculation unit uses this relationship to calculate the number of occupants on a floor for which elevator operation data cannot be collected. In this case, the number of occupants can be calculated more accurately even on floors for which elevator operation data cannot be collected. For example, the occupancy count calculation unit may use the more accurate method of calculating the number of occupants from the correction processing data or the physical quantity. In this case, the method of calculating the number of occupants from elevator operation data can calculate the number of occupants more accurately even if the accuracy of the number of occupants is not high.

[0009] Furthermore, the present invention provides a method for counting the number of people in a building, by having a processor execute a program stored in memory, to calculate the difference between the total number of people entering and the total number of people leaving on a floor of a building based on elevator operation data, to calculate correction processing data for correcting at least one of the number of people entering and the number of people leaving on the floor at a predetermined time based on the difference from the total, and to correct the number of people entering and the number of people leaving on the floor that have been calculated using the correction processing data based on the elevator operation data, thereby calculating the number of people occupying the floor.In this case, it is possible to provide a method for counting the number of people in a building, which can accurately correct errors in the number of people entering and the number of people leaving on a floor and accurately calculate the number of people occupying the floor. [Effects of the Invention]

[0010] The present invention aims to provide a number of people counting system and method that can accurately correct errors in the number of people entering and leaving a floor and accurately calculate the number of people present on a floor. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a block diagram showing the overall configuration of a building management system according to an embodiment of the present invention. [Figure 2] 1 is a flowchart illustrating an overview of the operation of a building management system according to the present embodiment. [Figure 3] 10 is a flowchart illustrating the processing of the offline processing unit in more detail. [Figure 4] 10 is a flowchart illustrating the processing of the online processing unit in more detail. [Figure 5] FIG. 1 is a diagram illustrating a user movement model. [Figure 6] 6 is a flowchart showing a process for determining the ratio shown in FIG. 5. [Figure 7]10 is a flowchart illustrating a process in which an offline processing unit calculates the difference between the total number of people entering a floor per day (Σ number of people entering) and the total number of people leaving (Σ number of people leaving). [Figure 8] 10 is a flowchart illustrating a process in which an offline processing unit calculates an index of an elevator usage state for a target floor from elevator operation data. [Figure 9] 10 is a flowchart illustrating a process performed by an offline processing unit to obtain correction processing data for correcting the number of people exiting a floor. [Figure 10] FIG. 10 is a diagram comparing calculation methods for each movement model, in which correction terms are calculated as correction processing data as in FIG. 9, and in which correction coefficients are calculated. [Figure 11] 10(a) and 10(b) are diagrams showing time-series data of the number of people entering, leaving, and present before correction. [Figure 12] 10(a) to 10(c) are diagrams showing the correction term for the number of people leaving, and the time series data of the number of people entering, leaving, and present after correction. [Figure 13] FIG. 10 is a diagram showing an example of measurement data relating to a person's activity state. [Figure 14] This is a flowchart illustrating a method for calculating the number of people in a room based on measurement data related to people's activity status when a method of calculating the number of people by obtaining correction processing data from elevator operation data results in a large error. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. <Overall explanation of building management system 1> FIG. 1 is a block diagram showing the overall configuration of a building management system 1 according to this embodiment. The building management system 1 of this embodiment is an example of an occupancy counting system, and calculates the number of people occupying each floor of a building based on elevator operation data. Here, the building is not particularly limited as long as it has multiple floors and is equipped with elevators to transport people to each floor, and examples of such buildings include office buildings and apartment buildings.

[0013] The building management system 1 is a computer device, such as a server computer. However, it is not limited to this and may be a personal computer (PC), a mobile computer, a smartphone, a tablet, or the like. It may also be a cloud server that operates on the cloud.

[0014] The building management system 1 includes a processor such as a CPU (Central Processing Unit) as a computing means, and a main memory as a storage means. The processor executes various software such as an OS (operating system) and applications (application software). The main memory is a storage area for storing various software and data used to run the software. The building management system 1 also includes storage such as an HDD (Hard Disk Drive) or SSD (Solid State Drive) as an auxiliary storage device, and a communication interface for communicating with the outside. The system may also include input devices such as a mouse and keyboard, and output devices such as a display.

[0015] As shown in the figure, the building management system 1 includes an entry / exit number calculation unit 10, an offline processing unit 20, and an online processing unit 30. The entrance number / exit number calculation unit 10 calculates the entrance number, which is the number of people who got off the elevator on each floor, based on the elevator operation data. Also, the entrance number / exit number calculation unit 10 calculates the exit number, which is the number of people who got on the elevator on each floor, based on the elevator operation data. The offline processing unit 20, which will be described in more detail later, calculates the difference between the total number of people entering and the total number of people leaving on a floor of a building based on elevator operation data, and calculates correction processing data that corrects at least one of the number of people entering and leaving on a floor at a specified time based on the difference from the total. The online processing unit 30, which will be described in more detail later, is an example of a number of people calculation unit, and based on elevator operation data, corrects the number of people entering and exiting a floor, for which correction processing data has been calculated, to calculate the number of people occupying a floor.

[0016] The entry / exit number calculation unit 10, the offline processing unit 20 and the online processing unit 30 will be described in more detail below. The entry and exit number calculation unit 10 includes an elevator data input unit 11, an elevator data storage DB (database) 12, a measurement data input unit 13 related to the activity state of people on the floor, a measurement data storage DB (database) 14 related to the activity state of people, an identification unit 15 of data available for the floor, and a floor entry and exit number calculation unit 16.

[0017] The elevator data input unit 11 receives elevator data from an elevator system that manages elevator operation. The elevator data is, for example, elevator operation data. The elevator operation data includes, for example, the floor where the elevator stopped and the weight of the elevator car relative to the time. The elevator data storage DB 12 stores elevator operation data input to the elevator data input unit 11.

[0018] Time series data of physical quantities related to the number of people on the floor is input to the measurement data input unit 13 related to the activity state of people on the floor. In this embodiment, the physical quantities related to the number of people on the floor are measurement data related to the activity state of people. As will be described in detail later, the measurement data related to the activity state of people includes, for example, data on power consumption and toilet usage status. The measurement data related to the activity state of people is input from, for example, a BEMS (Building Energy Management System), an environmental sensor, a toilet sensor, a water and sewage sensor, etc. The measurement data storage DB 14 relating to the activity state of people stores data input to the measurement data input unit 13 relating to the activity state of people on the floor.

[0019] The floor-related available data identification unit 15 identifies floors for which elevator data is available and floors for which elevator data is unavailable. Floors for which elevator data is unavailable are floors for which necessary data cannot be collected from the elevators, for example, because the elevators are old or made by a different elevator manufacturer.

[0020] The floor entry / exit number calculation unit 16 calculates the number of entry persons (t) and the number of exit persons (t) at a given time (t) from the elevator data. These can be calculated from the elevator car weight included in the elevator data. That is, by determining the weight per person in advance, the number of entry persons (t) and the number of exit persons (t) can be calculated based on the change in the elevator car weight at each floor, which can be detected by the opening and closing of the elevator car door. In this case, on floors close to the lobby floor, passengers often use the stairs to go down. Therefore, when looking at the entire day, the relationship between the total number of entry persons (Σ entry persons) and the total number of exit persons (Σ exit persons) is usually Σ entry persons > Σ exit persons. This difference is the error caused by using the stairs instead of the elevator. In this case, the number of exit persons (t) must be corrected to accurately calculate the number of people present. On the other hand, for example, if the same tenant occupies multiple floors, Σ entry persons < Σ exit persons may occur because people travel between floors by stairs. In this case, to accurately calculate the number of people present, it is necessary to correct the number of people entering (t). That is, if the total number of people leaving is smaller than the total number of people entering, the number of people leaving (t) is corrected, and if the total number of people entering is smaller than the total number of people leaving, the number of people entering (t) is corrected. As will be described in detail later, the offline processing unit 20 calculates correction processing data for making these corrections. The correction processing data is a correction term or correction coefficient for making these corrections.

[0021] The offline processing unit 20 includes a building floor information DB (database) 21, a building user inter-floor movement model DB (database) 22, a floor user movement model setting unit 23 for each floor, and an elevator usage status index calculation unit 24. The offline processing unit 20 also includes a difference calculation unit 25 for the number of people entering and leaving a floor, a correction processing data calculation unit 26 for the time series data of the number of people entering and leaving a floor, a time series data calculation unit 27 for the number of people occupying a floor, and a conversion coefficient calculation unit 28 for converting measurement data related to human activity into the number of people occupying a room.

[0022] The building floor information DB21 stores information about the floor configuration of the building and the purpose of each floor. The purpose information for each floor includes, for example, whether it is a lobby floor, a dining floor, or a shared floor (lounge floor). These floors have a higher volume of people moving about than general floors. In other words, people move mainly around these floors. For example, it is thought that many people enter the building from the lobby floor in the morning, move to their work floor, return to the lobby floor in the evening, and then leave the building. It is also thought that many people move from their work floor to the dining floor in the afternoon, and then return to the original floor. Information about the purpose of each floor can be used to obtain information about floors with a high volume of people moving about.

[0023] The building user inter-floor movement model DB22 stores multiple movement models that are models of people moving between floors. As will be described in detail later, in this embodiment, the movement models provided are: a) a model that uses the elevator and not the stairs, b) a model that uses the elevator when going up and the stairs when going down, c) a model that uses the stairs when the elevator is crowded and the elevator at other times, and d) a model that uses the stairs and not the elevator.

[0024] The floor user movement model setting unit 23 for each floor changes and sets the proportion of these movement models to be applied for each floor. Specifically, the floor user movement model setting unit 23 for each floor sets a higher proportion of models that use stairs rather than elevators for lower floors, for example. In other words, since it is thought that the lower the floor, the more people will move to the lobby floor by stairs rather than using the elevator, or move from the lobby floor to their destination floor without using the elevator, the setting is made to suit the actual situation.

[0025] The elevator usage status index calculation unit 24 calculates time series data of the number of elevator users for each floor and time series data of the average elevator waiting time for each floor as indexes representing the usage status of the elevator. These can be calculated based on the elevator data.

[0026] The floor entry and exit number difference calculation unit 25 is an example of a number difference calculation unit, and calculates the difference between the total number of entry people (Σ entry number) and the total number of exit people (Σ exit number) on a floor of a building based on elevator operation data. The total is, for example, the total number of entry people and exit people per day. Here, the difference = |Σ entry number - Σ exit number|.

[0027] The correction processing data calculation unit 26 for the time-series data of the number of people entering and leaving a floor is an example of a correction processing data calculation unit, and calculates correction processing data for correcting at least one of the number of people entering and leaving a floor at a predetermined time based on the difference from the total. Specifically, the correction processing data calculation unit 26 for the time-series data of the number of people entering and leaving a floor allocates the difference calculated by the difference calculation unit 25 between the number of people entering and leaving a floor to each movement model in the ratio set by the floor user movement model setting unit 23 for each floor. Then, the correction processing data calculation unit 26 for the time-series data of the number of people entering and leaving a floor calculates correction processing data for correcting the number of people entering or leaving a floor at a predetermined time (t). The actual processing will be described later.

[0028] The time series data calculation unit 27 for the number of people on the floor calculates the number of people on the floor at a specified time (t) as data necessary to calculate the conversion coefficient in the conversion coefficient calculation unit 28, which converts measurement data related to human activity into the number of people. The conversion coefficient calculation unit 28 converts the measurement data related to human activity into the number of people in the room, and calculates a conversion coefficient for converting the measurement data related to human activity into the number of people in the room. For example, if the measurement data related to human activity is the power consumption (kWh) of office automation equipment, the power consumption of office automation equipment per person (kWh / person) can be calculated as a conversion coefficient from the power consumption of office automation equipment divided by the number of people in the room. It is preferable to select floors for which elevator operation data can be acquired and which have similar conditions to floors for which elevator operation data cannot be collected, such as having the same tenant type or shape.

[0029] The online processing unit 30 includes a correction processing unit 31 for the number of people entering or leaving the floor, a floor occupancy number calculation unit 32, and a floor occupancy number calculation unit 33 based on measurement data relating to people's activity status.

[0030] The correction processing unit 31 for the number of people entering or leaving a floor corrects the number of people entering or leaving a floor for which correction processing data has been calculated, based on elevator operation data. This correction processing data is calculated by the correction processing data calculation unit 26 for the time-series data of the number of people entering and leaving a floor. Which of the number of people entering or leaving is corrected is determined by the result of the correction processing by the offline processing unit 20. That is, if the correction processing data calculation unit 26 for the time-series data of the number of people entering and leaving a floor calculates correction processing data for the total number of people entering, the correction processing unit 31 for the number of people entering or leaving a floor uses this correction processing data to correct the number of people entering. Also, if the correction processing data calculation unit 26 for the time-series data of the number of people entering and leaving a floor calculates correction processing data for the total number of people leaving, the correction processing unit 31 for the number of people entering or leaving a floor uses this correction processing data to correct the number of people leaving.

[0031] The floor occupancy calculation unit 32 calculates the number of occupants on each floor for which elevator data is available, based on the corrected number of people entering and leaving. The number of people entering or leaving each floor acquired by the correction processing unit 31 for the number of people entering or leaving each floor is, for example, newly acquired, and more specifically, is, for example, the current number. In this case, the correction processing unit 31 for the number of people entering or leaving each floor corrects the current number of people entering or leaving each floor. In addition, the floor occupancy number calculation unit 32 calculates the current (real-time) number of people occupying each floor based on the elevator data.

[0032] The floor occupancy calculation unit 33 from the measurement data related to human activity calculates the number of occupants on floors where elevator operation data cannot be collected using time-series data of physical quantities related to the number of occupants on the floor. More specifically, the conversion coefficient calculation unit 28, which converts the measurement data related to human activity to the number of occupants, calculates the relationship between the number of occupants and the physical quantities related to the number of occupants on floors where elevator operation data can be collected, and the floor occupancy calculation unit 33 from the measurement data related to human activity uses this relationship to calculate the number of occupants on floors where elevator operation data cannot be collected. As described above, this embodiment uses measurement data related to human activity. That is, the floor occupancy calculation unit 33 from the measurement data related to human activity converts the acquired measurement data related to human activity into the number of occupants using the conversion coefficient calculated by the time-series data calculation unit 27 of the number of occupants on the floor. The acquired measurement data related to human activity is, for example, newly acquired, and more specifically, is, for example, current data. In this case, the floor occupancy calculation unit 33, which calculates the number of people occupying each floor from measurement data related to people's activity status, calculates the number of people occupying each floor at the current time (real time) based on measurement data related to people's activity on floors where elevator operation data cannot be collected.

[0033] The number of people on each floor calculated by floor occupancy calculation unit 32 and floor occupancy calculation unit 33 based on measurement data related to people's activity status is output to the building management / operation department. The building management / operation department uses the calculated number of people on each floor as data for efficiently operating, for example, air conditioning and lighting equipment, cleaning, security, etc.

[0034] <Explanation of the operation of Building Management System 1> FIG. 2 is a flowchart outlining the operation of the building management system 1 according to this embodiment. 2 shows an outline of the processing of the offline processing unit 20 and the online processing unit 30. First, it is determined whether it is time to execute offline processing (S201). For example, the execution time is determined based on whether it is a predetermined time (e.g., 11:00 PM) on the day before the day on which processing is to be performed by the online processing unit 30. The execution time may be determined as one time for processing to be performed every day, or the day and time for processing may be changed taking into account the day of the week, seasonal variations, etc. As a result, if it is time to execute offline processing (Yes in S201), the loop process of S202 to S208 is executed as offline processing by the offline processing unit 20 for each floor of the target building. On the other hand, if it is not time to execute offline processing (No in S201), the process proceeds to S209.

[0035] In the loop process of S202 to S208, the offline processing unit 20 first sets the target floor (S203). Next, the offline processing unit 20 determines whether elevator operation data is available for the target floor (S204). As a result, if the elevator operation data is available (Yes in S204), the offline processing unit 20 calculates correction processing data for the number of people entering and leaving the elevator operation data (S205). Specifically, the correction processing data calculation unit 26 for the time-series data of the number of people entering and leaving the floor calculates correction processing data for correcting the number of people leaving the floor at a predetermined time (t). Furthermore, the offline processing unit 20 calculates a conversion coefficient for converting the measurement data relating to the activity state of people on the floor into the number of people present (S206). Specifically, the conversion coefficient calculation unit 28 for converting the measurement data relating to people's activity into the number of people present calculates the conversion coefficient for converting the measurement data relating to people's activity into the number of people present.

[0036] After S206, the process proceeds to S207. Also, if elevator operation data is not available in S204 (No in S204), the process proceeds to S207.

[0037] In S207, the offline processing unit 20 determines whether or not all floors / areas to be managed have been processed. As a result, if all floors / areas to be managed have been processed (Yes in S207), the loop process of S202 to S208 ends and the process proceeds to S209. On the other hand, if all floors / areas to be managed have not been processed (No in S207), the process returns to S203 and other target floors are set.

[0038] In S209, it is determined whether or not it is the day of the control. If it is the day of the control (Yes in S209), the loop process of S210 to S216 is executed as online processing for each floor of the target building by the online processing unit 30. On the other hand, if it is not the day of the control (No in S209), the series of processes ends.

[0039] In the loop process of S210 to S216, the online processing unit 30 first sets the target floor (S211). Next, the online processing unit 30 determines whether elevator operation data is available for the target floor (S212). As a result, if elevator operation data is available (Yes in S212), the online processing unit 30 calculates the number of people entering, leaving, and occupancy based on the elevator operation data (S213). Specifically, the floor number of people entering or leaving correction processing unit 31 corrects the acquired number of people entering and leaving for each floor that have been used to calculate the correction processing data. Then, the floor occupancy number calculation unit 32 calculates the number of people occupying each floor based on the corrected number of people entering and leaving. On the other hand, if elevator operation data is not available (No in S212), the online processing unit 30 calculates the number of people on the floor using measurement data related to the activity state of people (S214). Specifically, the floor occupancy number calculation unit 33, which calculates the number of people on the floor from measurement data related to the activity state of people, calculates the number of people on the floor using time-series data of physical quantities related to the number of people on the floor. After S213 and S214, the process proceeds to S215. In S215, the online processing unit 30 determines whether or not all floors / areas to be managed have been processed. As a result, if all floors / areas to be managed have been processed (Yes in S215), the loop process of S210 to S216 ends. On the other hand, if all floors / areas to be managed have not been processed (No in S215), the process returns to S211 and other target floors are set.

[0040] FIG. 3 is a flowchart illustrating the processing of the offline processing unit 20 in more detail. Here, the offline processing unit 20 performs a loop process of S301 to S313 for each floor of the target building. In this loop process, the offline processing unit 20 first sets the target floor (S302).

[0041] The following processing of S303 to S309 is a detailed explanation of the processing of S205 in Fig. 2. That is, in S303 to S309, the offline processing unit 20 calculates correction processing data for the number of people entering and leaving with respect to the elevator operation data. The offline processing unit 20 determines whether elevator operation data is available for the target floor (S303). As a result, if elevator operation data is available (Yes in S303), the offline processing unit 20 sets a user movement model for the target floor (S304) based on the floor position information for the target floor and the building user inter-floor movement model DB 22. This is processing performed by the floor user movement model setting unit 23 for each floor described above. Furthermore, the offline processing unit 20 calculates an index of elevator usage status for the target floor from the elevator operation data (S305). This is a process performed by the elevator usage status index calculation unit 24 described above, and the index of elevator usage status is time-series data of the number of elevator users for each floor, the average elevator waiting time for each floor, etc. Next, the offline processing unit 20 acquires (past) elevator operation data for the target floor from the elevator data storage DB 12 (S306). Next, the offline processing unit 20 calculates the number of people entering and leaving the target floor (S307). Then, the offline processing unit 20 calculates the difference between the total number of people entering the floor per day (Σ number of people entering) and the total number of people leaving (Σ number of people leaving) (S308). This is the processing performed by the above-mentioned floor number of people entering and number of people leaving difference calculation unit 25. Furthermore, the offline processing unit 20 calculates correction processing data for the number of people entering or leaving the floor based on 1) the user movement model, 2) the indicator of the elevator usage state, and 3) the difference between the total number of people entering and the total number of people leaving the floor (S309). This is the processing performed by the correction processing data calculation unit 26 described above.

[0042] The following processing in S310 to S311 is a detailed explanation of the processing in S206 in Fig. 2. That is, in S310 to S311, the offline processing unit 20 calculates a conversion coefficient for converting measurement data relating to the activity state of people on the floor into the number of people present. The offline processing unit 20 calculates the number of people present on the target floor based on the measurement data relating to the activity state of people on the floor (S310). Next, the offline processing unit 20 calculates a conversion coefficient for converting the measurement data related to the activity state of people on the target floor into the number of people present. This can be calculated using the formula: conversion coefficient = number of people present / measurement data value (S311). This is processing performed by the above-mentioned unit 27 for calculating the time-series data of the number of people present on the floor and the unit 28 for calculating the conversion coefficient for converting the measurement data related to people's activity into the number of people present.

[0043] After S311, the process proceeds to S312. Also, if elevator operation data is not available in S303 (No in S303), the process proceeds to S312.

[0044] In S312, the offline processing unit 20 determines whether all floors / areas to be managed have been processed. As a result, if all floors / areas to be managed have been processed (Yes in S312), the loop process of S301 to S313 ends. On the other hand, if all floors / areas to be managed have not been processed (No in S312), the process returns to S302 and other target floors are set.

[0045] FIG. 4 is a flowchart illustrating the processing of the online processing unit 30 in more detail. Here, the online processing unit 30 performs a loop process of S401 to S413 for each floor of the target building. In this loop process, the online processing unit 30 first sets the target floor (S402). Next, the online processing unit 30 determines whether elevator operation data is available for the target floor (S403). As a result, if the elevator operation data is available (Yes in S403), the online processing unit 30 inputs the elevator operation data for the target floor to the elevator data input unit 11 (S404). Furthermore, the online processing unit 30 stores the elevator operation data in the elevator data storage DB 12 (S405). Next, the floor entry / exit number calculation unit 16 calculates the number of entry and exit numbers for the target floor using the latest operation data for the day (S406). Furthermore, the online processing unit 30 corrects the number of people entering or leaving based on the correction processing data calculated by the offline processing unit 20 (S407). This is processing performed by the correction processing unit 31 for the number of people entering or leaving the floor described above. Then, the online processing unit 30 calculates the number of people occupying the target floor (S408). This is a process performed by the floor occupancy number calculation unit 32 described above. After S408, the process proceeds to S412.

[0046] Also, if elevator operation data is not available in S403 (No in S403), the online processing unit 30 inputs measurement data related to the activity status of people on the target floor to the measurement data input unit 13 related to the activity status of people on the floor (S409). Furthermore, the online processing unit 30 stores the measurement data relating to the activity state of people on the target floor in the measurement data storage DB 14 relating to the activity state of people (S410). Then, the online processing unit 30 calculates the number of people occupying the floor from the measurement data using a conversion coefficient that converts the measurement data related to the activity state of people on the target floor into the number of people occupying the floor (S411). This is a process performed by the above-mentioned floor occupancy number calculation unit 33 based on the measurement data related to the activity state of people. Furthermore, the online processing unit 30 determines whether or not all floors / areas to be managed have been processed (S412). As a result, if all floors / areas to be managed have been processed (Yes in S412), the loop process of S401 to S413 ends. On the other hand, if all floors / areas to be managed have not been processed (No in S412), the process returns to S402 and other target floors are set.

[0047] FIG. 5 is a diagram illustrating a user movement model. Figure 5 explains the user movement model using a table that corresponds to the movement model number, the movement model of the building user floor, the features for detecting moving users, the detection method for moving users, and the proportion of people. Here, the user movement models shown are the above-mentioned a) model in which the elevator is used and the stairs are not used (No. 1), b) model in which the elevator is used when going up and the stairs when going down (No. 2), c) model in which the stairs are used when the elevator is crowded and the elevator is used at other times (No. 3), and d) model in which the elevator is not used and the stairs are used (No. 4).

[0048] a) In the case of model (No. 1) where people do not use stairs but use elevators, it is shown that elevator operation data can be used as a feature for detecting moving users. It also shows that elevator operation data can be used as a method for detecting moving users. b) In the case of model No. 2, where people use the elevator to go up and the stairs to go down, a feature of detecting moving passengers is that movement on stairs cannot be detected using elevator data, so it is shown to be reflected in the difference between the number of people entering and exiting calculated using elevator operation data.In addition, it is shown that a method of detecting moving passengers is to estimate based on the difference between the number of people entering and exiting calculated using elevator operation data. c) Model No. 3, in which people use the stairs when the elevator is crowded and the elevator at other times, is the same as model No. 2, in which people use the elevator when going up and the stairs when going down. d) In the case of model 4, where users do not use elevators but use stairs, the model shows that users cannot be detected using elevator operation data. Also, the model shows that users can be detected by assuming that a certain percentage of users move between floors, especially on floors close to the lobby floor, exist.

[0049] The percentages indicate the proportion to which these movement models are applied for each floor. Here, the application rates for each movement model are 70%, 15%, 10%, and 5%. Note that these percentages vary depending on the floor. This is because the proportion of elevator users increases on higher floors, and the proportion of stair users increases on lower floors. This allows us to adapt to the reality that people tend to use stairs and not elevators on lower floors. In this embodiment, the difference between the total number of people entering each floor (ΣNumber of people entering) and the total number of people leaving (ΣNumber of people leaving) is allocated to each movement model at this rate, and the number of people entering and leaving is corrected. This makes it possible to more accurately estimate how the difference, which is an error, occurred, and to calculate more accurate correction processing data.

[0050] FIG. 6 is a flowchart showing the process of determining the ratio shown in FIG. Here, the offline processing unit 20 performs a loop process of S601 to S611 for each floor of the target building. In this loop process, the offline processing unit 20 first sets the target floor (S602). Next, the offline processing unit 20 determines whether elevator operation data is available for the target floor (S603). As a result, if elevator operation data is available (Yes in S603), the offline processing unit 20 acquires floor information of the target floor (S604). This corresponds to, for example, location information of the target floor, the positional relationship between the target floor and the lobby floor or dining floor, etc. Next, the offline processing unit 20 determines whether the difference between the total number of people entering (Σnumber of people entering) and the total number of people leaving (Σnumber of people leaving) for the target floor is equal to or greater than a predetermined value (S605). As a result, if the value is equal to or greater than the predetermined value (Yes in S605), the offline processing unit 20 sets a) the ratio of models that do not use stairs but use elevators based on the floor information (S606). Furthermore, the offline processing unit 20 sets the ratio of models that use elevators to go up and stairs to go down based on the floor information (b) (S607). Furthermore, the offline processing unit 20 sets, based on the floor information, c) the proportion of models that use the stairs when the elevator is crowded and use the elevator otherwise (S608). Furthermore, the offline processing unit 20 sets d) the proportion of models that do not use elevators but use stairs, based on the floor information (S609).

[0051] After S609, the process proceeds to S610. Also, if elevator operation data is not available in S603 (No in S603), the process proceeds to S610.

[0052] In S610, the offline processing unit 20 determines whether all floors / areas to be managed have been processed. As a result, if all floors / areas to be managed have been processed (Yes in S610), the loop process of S601 to S611 ends. On the other hand, if all floors / areas to be managed have not been processed (No in S610), the process returns to S602 and other target floors are set.

[0053] FIG. 7 is a flowchart illustrating the process performed by the offline processing unit 20 to calculate the difference between the total number of people entering a floor per day (Σnumber of people entering) and the total number of people leaving (Σnumber of people leaving). This is a more detailed explanation of the processes in S306 to S308 in FIG. First, past elevator operation data is input to the offline processing unit 20 (S701). Next, the offline processing unit 20 calculates time series data of the number of people entering and leaving for each floor from the past elevator operation data (S702). Furthermore, the offline processing unit 20 calculates the total number of people entering each floor (per day) and the total number of people leaving each floor (per day) (S703). The offline processing unit 20 then calculates the difference between the total number of people entering each floor (per day) and the total number of people leaving (per day).The offline processing unit 20 then stores which of the total number of people entering and the total number of people leaving is larger for each day, and the calculated difference.

[0054] FIG. 8 is a flowchart illustrating the process performed by the offline processing unit 20 to calculate an index of the elevator usage state for a target floor from elevator operation data. This is a more detailed explanation of the process of S305 in FIG. First, past elevator operation data is input to the offline processing unit 20 (S801). Next, the offline processing unit 20 calculates the number of elevator users and average waiting time for each floor, time, and direction (upward and downward) from the past elevator operation data (S802). Furthermore, the offline processing unit 20 outputs the result calculated in S802 to the correction processing data calculation unit 26 for the time series data of the number of people entering and leaving the floor (S803).

[0055] FIG. 9 is a flowchart illustrating the process performed by the offline processing unit 20 to obtain correction processing data for correcting the number of people leaving the floor. This is a more detailed explanation of the process of S309 in Fig. 3. Fig. 9 explains the case where a correction term is obtained as correction processing data. Here, the offline processing unit 20 performs a loop process of S901 to S909 for each floor of the target building. In this loop process, the offline processing unit 20 first sets the target floor (S902). Next, the offline processing unit 20 determines whether elevator operation data is available for the target floor (S903). As a result, if elevator operation data is available (Yes in S903), the offline processing unit 20 inputs information indicating which is larger, the total number of people entering or the total number of people leaving, for the target floor (S904).

[0056] Next, the offline processing unit 20 calculates a correction term for the model b) where people use the elevator to go up and the stairs to go down (S905). Specifically, the correction term is calculated by allocating the difference between the time series data of the total number of people entering and the total number of people leaving, whichever is smaller, according to the above ratio. For example, if the total number of people leaving is smaller, the correction term (t) for the number of people leaving at time (t) in this model is calculated using the number of people leaving (t) before correction at time (t), the total number of people leaving, and the ratio, using the following formula (1):

[0057] Correction term for number of people leaving (t) = (difference × (number of exits before correction (t) / total number of exits)) × ratio ... (1)

[0058] Next, the offline processing unit 20 calculates a correction term for the model c) where people use the stairs when the elevators are crowded and use the elevators at other times (S906). Specifically, the correction term (t) for the number of people exiting at time (t) in this model is calculated using the number of people using the elevator at time (t) (t), the total number of people using the elevators, and the ratio, using the following formula (1).

[0059] Correction term for number of people leaving (t) = (difference × (number of elevator users (t) / total number of elevator users)) × ratio…(2)

[0060] Furthermore, the offline processing unit 20 calculates d) a correction term for a model in which elevators are not used and stairs are used (S907). Specifically, the correction term (t) for the number of people entering and the correction term (t) for the number of people leaving at time (t) of this model are calculated by using the number of people entering before correction (t) and the number of people leaving before correction (t), respectively, and multiplying these by a ratio as shown in the following equations (3) and (4).

[0061] Correction term for number of visitors (t) = number of visitors before correction (t) × ratio … (3) Correction term for number of exits (t) = number of exits before correction (t) × ratio ... (4)

[0062] After S907, the process proceeds to S908. Also, if elevator operation data is not available in S903 (No in S903), the process proceeds to S908.

[0063] In S908, the offline processing unit 20 determines whether all floors / areas to be managed have been processed. As a result, if all floors / areas to be managed have been processed (Yes in S908), the loop process of S901 to S909 ends. On the other hand, if all floors / areas to be managed have not been processed (No in S908), the process returns to S902 and the setting of other target floors is performed.

[0064] FIG. 10 is a diagram comparing the calculation methods for each movement model when calculating correction terms as correction processing data as in FIG. 9 and when calculating correction coefficients. a) For the model (No. 1) in which people do not use stairs but use elevators, there are no correction terms or correction factors. b) In model No. 2, where elevators are used for going up and stairs for going down, the correction term can be calculated using the above formula (1). The correction coefficient for this model at time (t) is calculated using the following formula (5).

[0065] Correction coefficient for number of exits (t) = (difference / total number of exits) × ratio ... (5)

[0066] c) In model No. 3, where people use the stairs when the elevator is crowded and the elevator at other times, the correction term can be calculated using equation (2) above. Also, the correction coefficient for this model at time (t) is calculated using equation (6) below.

[0067] Correction coefficient for number of people leaving (t) = (difference / total number of people using elevators) x ratio ... (6)

[0068] d) In model No. 4, where elevators are not used and stairs are used, the correction terms can be calculated using the above equations (3) and (4). In addition, the correction coefficients at time (t) for this model are calculated using the following equations (7) and (8).

[0069] Correction coefficient for the number of visitors (t) = ratio ... (7) Correction coefficient for number of people leaving (t) = ratio … (8)

[0070] The correction processing data used by the online processing unit 30 to correct the number of people entering and leaving is the sum of the correction processing data calculated for each movement model (hereinafter, this may be referred to as "model correction processing data"). When a correction term is used as the correction processing data, the correction term (t) for the number of people leaving is the correction term (t) for the number of people leaving in equation (1) + the correction term (t) for the number of people leaving in equation (2) + the correction term (t) for the number of people leaving in equation (4). When a correction coefficient is used as the correction processing data, the correction coefficient (t) for the number of people leaving is the correction coefficient (t) for the number of people leaving in equation (5) + the correction coefficient (t) for the number of people leaving in equation (6) + the correction coefficient (t) for the number of people leaving in equation (8).

[0071] This can also be said to be that the offline processing unit 20 sets a plurality of movement models, calculates model correction processing data as correction processing data for each movement model, and obtains the correction processing data from the model correction processing data. It can also be said that the offline processing unit 20 determines the ratio at which multiple movement models are applied, allocates the above-mentioned difference according to the determined ratio, calculates model correction processing data from the allocated difference, and uses the sum of the calculated model correction processing data as the correction processing data.

[0072] 11(a) and 11(b) are diagrams showing time-series data of the number of people entering, leaving, and present before correction, where the horizontal axis represents time and the vertical axis represents the number of people. These are time-series data that show the changes in the number of people entering, leaving, and present in the room for each hour between 0:00 and 24:00 on one day. Figure 11(a) shows the time series data of the number of people entering and leaving a certain floor before correction, and Figure 11(b) shows the number of people present calculated based on the time series data in Figure 11(a). In this case, the number of people present during late night hours should be 0. However, since the total number of people entering (Σ Number of people entering) > the total number of people leaving (Σ Number of people leaving), the number of people present will not be 0 even during late night hours, and the difference between the total number of people entering and the total number of people leaving will be counted as the number of people present. Note that on floors with night shifts, the number of people present will not necessarily be 0 even during late night hours, so in such cases, this difference is adjusted to a value that takes into account the number of people on the night shift, and the above movement model is applied.

[0073] 12(a) to 12(c) are diagrams showing the time series data of the number of people entering, leaving, and present after correction of the number of people leaving, where the horizontal axis represents time and the vertical axis represents the number of people. These are time series data that show the changes in the correction term for the number of people leaving, the number of people entering, the number of people leaving, and the number of people present, for each hour between 0:00 and 24:00 on one day. Of these, Figure 12(a) shows the correction term obtained by allocating the difference shown in Figure 11(b) according to the number of people exiting. Here, the correction term for model b) where people use the elevator to go up and the stairs to go down is shown. Here, it shows that correction terms were generated for 12:00 and 18:00. In other words, Figure 12(a) shows that a correction term (t) for the number of people exiting was generated for the cases of 12:00 and 18:00, where time (t) is taken as the time. Figure 12(b) shows the time series data of the number of people entering and leaving when the number of people leaving has been corrected using this correction term. Compared to Figure 11(a), this correction has resulted in an increase in the number of people leaving at 12:00 and 18:00. In other words, Figure 12(b) shows that the number of people leaving has been corrected for the times (t) of 12:00 and 18:00. Figure 12(c) shows the number of people present calculated based on the number of people entering and leaving after the correction in Figure 12(b). In this case, the difference shown in Figure 11(b) has been corrected and disappeared during the late-night hours, and the number of people present has become zero.

[0074] FIG. 13 is a diagram showing an example of measurement data relating to a person's activity state. Here, the relationship between the units of measurement data and the number of people present is shown for the measurement data related to the activity state of people shown in No. 1 to No. 7. Examples of measurement data related to people's activity include data on the power consumption of the power outlets to which personal office equipment on the target floor is connected (power consumption of "office outlets") (No. 1), data on the usage status of the toilets on the target floor (water consumption in the toilets, number of times each toilet room is used, etc.) (No. 2), data on the number of times people enter and exit the entrance and exit management system on the target floor (No. 3), measurement data on CO2 concentration on the target floor (No. 4), data on the amount of water used by the water supply and sewerage systems on the target floor (No. 5), measurement data on noise on the target floor (No. 6), and measurement data on the amount of waste on the target floor (No. 7). These data increase in accordance with the number of people occupying the room, so there is a correlation between these measurement data and the number of people occupying the room. Therefore, the number of people occupying the room can be calculated from these measurement data.

[0075] The online processing unit 30 can use either a method of calculating the number of people using correction processing data or a method of calculating the number of people from measurement data related to people's activity status, whichever is more accurate. For example, if there is a large error in the method of calculating the number of people by calculating correction processing data from elevator operation data, it is possible to use a method of calculating the number of people based on measurement data related to people's activity status.

[0076] FIG. 14 is a flowchart illustrating this case. Here, the online processing unit 30 performs a loop process of S1401 to S1408 for each floor of the target building. In this loop process, the online processing unit 30 first sets the target floor (S1402). Next, the online processing unit 30 determines whether or not elevator operation data is available for the target floor (S1403). As a result, if elevator operation data is available (Yes in S1403), it is determined whether the difference between the total number of people entering (per day) and the total number of people leaving (per day) for the target floor is greater than or equal to a predetermined value (S1404). If the difference is equal to or greater than the predetermined value (Yes in S1404), the online processing unit 30 determines that the calculation of the number of people using the elevator operation data has a large error, and calculates the number of people using the measurement data related to the activity state of people on the target floor (S1405). After this, the process proceeds to S1407. If the difference is less than the predetermined value (No in S1404), the process proceeds to S1406.

[0077] On the other hand, if elevator operation data is not available in S1403 (No in S1403), the process of calculating the number of people in the room using the elevator operation data described in the flowchart of Fig. 4 is executed (S1406). After this, the process proceeds to S1407.

[0078] In S1407, the online processing unit 30 determines whether all floors / areas to be managed have been processed. As a result, if all floors / areas to be managed have been processed (Yes in S1407), the loop process of S1401 to S1408 ends. On the other hand, if all floors / areas to be managed have not been processed (No in S1407), the process returns to S1402 and the setting of other target floors is performed.

[0079] <Explanation of effect> According to the building management system 1 described above, the offline processing unit 20 calculates, for example, the difference between the total number of people entering a building (Σ number of people entering) and the total number of people leaving a building (Σ number of people leaving) per day based on past elevator operation data. Furthermore, the offline processing unit 20 uses this difference to calculate correction processing data for correcting at least one of the errors between the number of people entering and the number of people leaving. The correction processing data is a parameter for correcting the number of people entering or leaving a floor at a given time (t), and is the correction term or correction coefficient described above. This makes it possible to calculate correction processing data that can more accurately correct the number of people entering or leaving. Furthermore, the offline processing unit 20 sets multiple movement models to model the movement of people between floors, calculates model correction processing data as correction processing data for each movement model, and calculates the correction processing data from the model correction processing data. In this case, by using a movement model that matches the actual movement of people and appropriately adjusting the application ratio, more accurate correction processing data can be calculated. The online processing unit 30 then uses this correction processing data to calculate, for example, the number of people present in a room in real time. In this case, correction is made not to the number of people present in the room, but to the number of people entering and leaving. This allows for accurate correction of errors in the number of people entering and leaving the floor, and allows for accurate calculation of the number of people present on the floor.

[0080] The number of people moving between floors and their movement patterns vary depending on the day of the week. For example, in buildings such as office buildings, the number of people coming in and out is often completely different between weekdays and holidays. Therefore, when the online processing unit 30 calculates the number of people present, it is preferable that the past elevator operation data input by the offline processing unit 20 be for the same day of the week. Similarly, the past elevator operation data may be selected taking into account seasonal variations such as holidays, Golden Week, and the New Year's holiday season.

[0081] <Explanation of how to count the number of people in a room> The processes performed by the building management system 1 are realized by the cooperation of software and hardware resources. Therefore, the processing performed by the above-mentioned building management system 1 can be considered to be a method of counting the number of people present in a room, in which the processor executes a program recorded in memory to determine the difference between the total number of people entering and the total number of people leaving a floor of a building based on elevator operation data, calculates correction processing data to correct at least one of the number of people entering and leaving the floor at a specified time based on the difference from the total, corrects the number of people entering and leaving the floor that has been calculated using the correction processing data based on the elevator operation data, and calculates the number of people present on the floor. In addition, the program running in the building management system 1 can be considered to be a program that enables a computer to perform the following functions: calculate the difference between the total number of people entering and the total number of people leaving a floor of a building based on elevator operation data, and calculate correction processing data that corrects at least one of the number of people entering and leaving a floor at a specified time based on the difference with the total; and correct the number of people entering and leaving a floor that has been calculated using the correction processing data based on the elevator operation data, and calculate the number of people occupying the floor.

[0082] The program for realizing this embodiment can be provided not only by communication means but also by being stored on a recording medium such as a CD-ROM.

[0083] Although the present embodiment has been described above, the technical scope of the present invention is not limited to the scope described in the above embodiment. It is clear from the claims that various modifications and improvements to the above embodiment are also included in the technical scope of the present invention. [Explanation of symbols]

[0084] 1... Building management system, 10... Entry and exit number calculation unit, 20... Offline processing unit, 21... Building floor information DB, 22... Building user movement model DB between floors, 23... Floor user movement model setting unit for each floor, 24... Elevator usage status index calculation unit, 25... Floor number of entry and exit number of entry and exit number of entry and exit, 26... Correction processing data calculation unit for time series data of entry and exit numbers of entry and exit ...

Claims

1. a number-of-people difference calculation unit that calculates the difference between the total number of people entering and the total number of people leaving a building floor based on elevator operation data; a correction processing data calculation unit that calculates correction processing data for correcting at least one of the number of people entering and leaving the floor at a predetermined time based on the difference from the total; a number-of-occupancy calculation unit that calculates the number of people occupying the floor by correcting the correction processing data among the number of people entering and exiting the floor based on the elevator operation data; A people counting system equipped with the above.

2. The number of people present in a room counting system described in claim 1, wherein the correction processing data calculation unit sets multiple movement models as models of people moving between floors, calculates model correction processing data as correction processing data for each of the movement models, and obtains the correction processing data from the model correction processing data.

3. The correction processing data calculation unit determines the ratio at which the multiple movement models are applied, allocates the difference according to the determined ratio, calculates the model correction processing data from the allocated difference, and uses the sum of the calculated model correction processing data as the correction processing data.

4. The occupancy counting system according to claim 3 , wherein the correction processing data calculation unit changes the ratio for each floor.

5. The number of people present in a room counting system described in claim 4, wherein when calculating correction processing data to correct the total number of people exiting, the correction processing data calculation unit uses as the movement model a model in which stairs are not used and elevators are used, a model in which elevators are used to go up and stairs to go down, a model in which stairs are used when the elevators are crowded and elevators are used at other times, and a model in which elevators are not used and stairs are used, and changes the ratio at which these movement models are applied for each floor.

6. The number of people counting system according to claim 5 , wherein the correction processing data calculation unit increases the proportion of models that use stairs and not elevators for lower floors.

7. The number of people present in a room counting system described in claim 1, wherein the correction processing data calculation unit calculates correction processing data to correct the number of people leaving when the total number of people leaving is smaller than the total number of people entering, and calculates correction processing data to correct the number of people entering when the total number of people entering is smaller than the total number of people leaving.

8. The number of people counting system according to claim 1, wherein the number of people calculation unit calculates the number of people on a floor where elevator operation data cannot be collected using time series data of physical quantities related to the number of people on the floor.

9. the correction processing data calculation unit calculates a relationship between the number of people in a room and a physical quantity related to the number of people in the room on a floor for which elevator operation data can be collected; 9. The system according to claim 8, wherein the number of people calculation unit uses this relationship to calculate the number of people on a floor for which elevator operation data cannot be collected.

10. The number of people counting system according to claim 8, wherein the number of people calculation unit adopts the more accurate method of calculating the number of people using the correction processing data or calculating the number of people from the physical quantity.

11. The processor executes the program stored in the memory. Based on elevator operation data, the difference between the total number of people entering and leaving a building floor is calculated. Based on the difference from the total, correction processing data is calculated to correct at least one of the number of people entering and leaving the floor at a predetermined time; Based on the elevator operation data, the number of people entering and leaving the floor, which has been calculated using the correction processing data, is corrected to calculate the number of people occupying the floor. How to measure the number of people in the room.

Images