0001]The invention, the EMS, and relates to the management method of an elevator, for example, relates to an elevator management system for managing a plurality of car which runs a plurality of floors interbed as a group.
BACKGROUND
[0002]Conventionally, the EMS of this kind, between the lowest floor of the top floor, so that the distance in the direction of gravity between a plurality of the car is equalized, thereby reciprocating the elevator car. However, for some of the car, the delay due to the passenger of a large number of passengers generated, stops can not be operated uniformly a plurality of car, for example, a plurality of cars in the first floor floor same time such as by, in the other floor, a situation such as the waiting time is longer has occurred.
[0003]
　Therefore, Patent Document 1, in order to improve the travel efficiency of the car, the invention controls the elevator management system so as to equalize the latency of the car at each floor is proposed. The present invention, while assuming that each of the plurality of car husband repeated operation between the lowest floor and the top floor, and set the moving direction and position of the car after a predetermined time, to fit thereto to, is that to navigate the car.
CITATION
Patent Document
[0004]
Patent Document 1: Japanese Patent No. 4139819
Summary of the Invention
Problems that the Invention is to Solve
[0005]
　In a conventional elevator control system, there is a problem that is wasted on travel of the car occur. Accordingly, the present invention provides an elevator control system which is adapted to navigate the car efficiently, and is intended to provide a management method of the elevator.
Means for Solving the Problems
[0006]
　In the present invention for solving the above problems, in the elevator management system for managing an elevator provided with a control device for operating over a car to a plurality of floors, comprising a management apparatus for managing a control device, the management device , destination floor designation information, and learns a receiving circuit for receiving the car call information, a memory for recording by accumulating received information of the reception circuit, based on the information recorded in the memory, the operation tendency of the car comprising a controller, and an output circuit for outputting the management information to the control device, the controller is predicted based on the results of learning, destination floor designated information from the received information after a predetermined time of the reception circuit, and the car call information and, based on the result of the prediction after the predetermined time, and formed so as to restrict the operation floor ranging from the cage management information, the control device, based on the management information And so as to control the operation of Rikago.
[0007]
　In the present invention also in the management method of an elevator for managing control device for operating over a car of an elevator into a plurality of floors by the management apparatus, the management unit, destination floor designation information, and receiving the car call information and, recording and accumulating the received information of the reception circuit, based on the received information, to learn the operation tendency of the car, it outputs the management information to the control device as a learning result, based on the learning result, the received information destination floor designation information after a predetermined time, and predicts the car call information, based on the result of the prediction, and formed so as to restrict the operation floor ranging from the cage management information, the control device, the management information and so as to control the operation of the car based on.
Effect of the invention
[0008]
　According to the present invention, an elevator management system so as to travel the car efficiently, and can be realized how to manage the elevator.
BRIEF DESCRIPTION OF THE DRAWINGS
[0009]
FIG. 1 is a block diagram showing a configuration of an elevator control system according to this embodiment.
FIG. 2 is a schematic diagram showing a principal part of the schematic structure of an elevator apparatus according to this embodiment.
3 is a diagram showing a service route of the elevator apparatus according to this embodiment.
4 is a conceptual diagram explaining the learning according to the present embodiment.
5 is a conceptual diagram illustrating the configuration of operational data table according to the present embodiment.
6 is a conceptual diagram showing a configuration of a learning data table according to the present embodiment.
7 is a flowchart illustrating a processing procedure of the operation data storage process.
8 is a flowchart illustrating a processing procedure of the operation data learning processing.
9 is a flowchart showing a processing procedure of service route determination process.
It is a flowchart showing the FIG. 10 operation instruction processing in the processing procedure.
11 is a flowchart showing a processing procedure in the service route correction process operation.
12 is a flowchart illustrating a processing procedure of the door opening when the service route correction process.
13 is a block diagram showing a configuration of an elevator control system according to another embodiment.
14 is a block diagram showing a configuration of an elevator control system according to another embodiment.
Is a block diagram showing a configuration of an elevator control system according to [15] Other embodiments.
16 is a block diagram showing a configuration of an elevator control system according to another embodiment.
DESCRIPTION OF THE INVENTION
[0010]
(1) Configuration of the EMS according to the present embodiment
　in Figure 1, 1 denotes an elevator management system according to this embodiment. The EMS 1 is configured with a management server 2 for managing a plurality of elevators 3. Management server 2 and a plurality of elevators 3 are connected via a communication path 19 such as an intranet.
[0011]
　Management server 2, the operation data of each elevator 3, acquired through the receiving circuit, to learn the operation status of each elevator 3 from the acquired operational data, the operation of each elevator 3, an output circuit by outputting the management information via a management apparatus that manages. The management server 2 is configured with a CPU (Central Processing Unit) 4, an auxiliary storage device 5 and the memory 6.
[0012]
　CPU4 is a processor (controller) which controls the management server 2 overall operation control. The auxiliary storage device 5, for example, a hard disk device, is composed of a non-volatile mass storage device, such as a SSD (Solid State Drive), is used for a long time storing programs and data. This part of the storage area where the auxiliary storage device 5 is provided is used as the operational data table TB10 and the learning data table TB20 later.
[0013]
　Memory 6 is composed of, for example, a volatile semiconductor memory, is also used as a work memory of the CPU 4, comprising the operation storage module 7, operation learning module 8, the path decision module 9 and routing module 10. The memory 6, operational data may be appropriately recorded cumulatively.
[0014]
　Each elevator 3, as shown in FIG. 2, the hoistway installed in a building, for example, between landings respectively provided from the first floor to the seventh floor of the hierarchy, and operated to raise and lower the elevator car 12 ing.
[0015]
　The elevator car 12 is attached to the other end of the main rope 13 which counterweight 14 is attached to one end. The main ropes 13 are wound around the hoisting machine 15. Hoist 15 is a lifting mechanism for vertically driving the elevator car 12, a machine room provided in the upper portion of the hoistway, the control unit (hereinafter for controlling the lifting operation of the car 12, this and the elevator control hereinafter) are installed along with 11.
[0016]
　Elevator controller 11 (FIG. 1) is a computer device that controls the operation of the car 12, the car 12 in accordance with the passenger operation (car call information) for call button 16 provided in the hall (2) controls the hoist 15 so as to lift the.
[0017]
　Elevator 3, it is difficult to determine whether or not to run the car 12 ride which path to which time zone, but inefficient, usually the elevator car 12 to the lowermost floor from the top floor It is operated as those that can be operated. In the present invention, it implements a learning function to make this determination to the elevator management system 1.
[0018]
　Next, a description will be given learning function implemented in the management server 2 in accordance EMS 1. The learning function can, for example, carrying out the deep learning.
[0019]
　Learning function of the elevator management system 1, car call information of each hierarchy, and / or, after receiving an operation on destination floor designated buttons for each elevator car 12 (destination floor specifying information), after a predetermined time (for example, riding car 12 learns the prediction to travel trend the operational state of the destination floor designated button call button 16 and the car 12 after 5 minutes) of the roadway as a period of one round trip.
[0020]
　Figure 3 shows an example of a learning function. Between neurons of adjacent layers (rows of 3) is the weighted (circle in Fig. 3), calculation is performed. This learning function, when the input number of the array dimensions of destination floors specified button call button 16 and the elevator car 12 in each layer is input from the input layer, a plurality of hidden layers and the output layer, the management server 2 performs a predetermined operation. Incidentally input layer input is one in Okonawaru hierarchy, the output layer is one in hierarchy output is performed. The layer hidden middle of the input layer and the output layer there is a plurality.
[0021]
　And as a result of the operation, the management server 2 outputs an input number of the array dimensions of destination floors specified button call button 16 and the elevator car 12 in each layer. The predetermined operation in the hidden layer, for example, a calculation using sigmoid function, hyperbolic tangent function, the activation function, such as a ramp function. Further the predetermined calculation in the output layer, for example, a calculation using the softmax function like.
[0022]
　In FIG 3, the management server 2, based on this learning function, called the elevator car 12 rises on the first floor, and, if the car 12 to the elevator car 12 and the descending increases at the second floor is called , it predicts whether called Gado car 12 in the next cycle.
[0023]
　Management server 2, in this case, in the next cycle, the elevator car 12 rises in the second floor is called, and predicts that the elevator car 12 is called a descending at the fourth floor.
[0024]
　In FIG. 3 represents the case where button is pressed "○", the case where button is not pressed by "×". In addition, in each hierarchy, "↑" the button of the case to call the elevator car 12 to rise in the hall, buttons if you get off the button of the case to call the elevator car 12 to fall in the hall, "↓", in the floor in the car 12 is represented by "→" the. In FIG. 3, since the car 12 is an example where only one, but the column of "→" is a column by column on each floor, or a plurality.
[0025]
　For prediction of FIG. 3, by deep learning, the management server 2 predicts the cage 12 is not called in the seventh floor of the fourth floor during the time it takes to one round trip. The management server 2 instructs the elevator controller 11 so as to navigate the car 12 in the service route for the purpose floor to the fourth floor as shown by a solid line in FIG. Specifically, the management server 2, after the waiting time for the passenger at the 4th floor, so as to reverse the traveling direction of the car 12, instructs the elevator controller 11. Note the dashed line in FIG. 4 shows a conventional service route, in the conventional operation the car 12 to ride up the seventh floor is the top floor to reach.
[0026]
　As a means for realizing the learning function as described above, as shown in FIG. 1, the memory 6 of the management server 2, operation storage module 7, operation learning module 8, the route determination module 9 and routing module 10 stored in the auxiliary storage device 5 of the management server 2, the operational data table TB10 and the learning data table TB20 is stored.
[0027]
　Travel storage module 7 is a program having a function of acquiring the operation data for example, every day from each of the elevators 3 of the elevator control device 11, and stores the obtained operation data into the operation data table TB10.
[0028]
　Travel learning module 8 performs one year to based on operational data acquired from the operation data table TB10 shown in FIG. 3 learning example every year, learning values ​​or the like of the weighting, as shown in FIG. 3 is a program to change the value necessary for. The operation learning module 8, the state of the destination floor designated button state and the elevator car 12 of the call button 16 in each layer was calculated as a learning result (learning result) is recorded in the learning data table TB20. Incidentally, each of the learning result is calculated for the combination of states of the any destination floor specification button state and the elevator car 12 of the call button 16 in each layer.
[0029]
　Path decision module 9 acquires the learning result from the learning data table TB20, a program for determining a service route of the car 12. Path decision module 9, from each of the learning result, derive a path can be omitted unnecessary to run the elevator car 12 is determined as a service route of the car 12 the path not passing through the route as the short path.
[0030]
　For example, when the learning result as input line and an output line in FIG. 6, the path decision module 9, in the input line, it can be seen that there is no need to pass through the 7 floors 2 even output line.
[0031]
　Therefore, the path decision module 9, the shortcut path not passing through 7 floors 2, determined as service route of the car 12. In the case omitted can route no route determination module 9 determines the travel path of the car 12 take the normal path for reciprocating the top floor from the lowest floor.
[0032]
　Routing module 10 is a program for correcting the travel path route determination module 9 is determined depending on the position and the traveling direction and the like of each of the car given from each of the elevators 3 of the elevator control device 11. For example, during travel of the car 12, within the service route of the elevator car 12 has not yet passed, and if the call button 16 is pressed, when the door of the car 12 is open, still occur the service route of the car 12 is corrected to ride like if there is no prediction, as management information the corrected service route, and transmits to the elevator control 11 the elevator 3.
[0033]
　In the case where the service route of the car 12 does not have to be corrected, routing module 10 transmits the determined service route path decision module 9 as it is to the elevator control 11 the elevator 3. The routing module 10 based on the determined service route path decision module 9 determines a car 12 that runs from the plurality of the car 12 more than one.
[0034]
　The operation data table TB10, as shown in FIG. 5, the state of the call buttons 16 in each layer (car call information) and state (destination floor specifying information) of a destination floor designated buttons for each elevator car 12, as operation data 5 minutes and stored for each (the time it takes to one round trip). It should be noted that "○" in FIG. 5, "×", "↑", "↓", "→" is the same meaning as in FIG. 3.
[0035]
　Similarly, the learning data table TB20, as shown in FIG. 6, when the input status of the destination floor designated button state and the elevator car 12 of the call button 16 at any respective hierarchies, 5 min (1 state of destination floor specification button state and the elevator car 12 of the call button 16 in each layer of the time) for reciprocating is stored as an output. It should be noted that "○" in FIG. 6, "×", "↑", "↓", "→" is the same meaning as in FIG. 3.
[0036]
(2) various processes by the management server
　will be described various processing executed in the management server 2 described above. In the following is described where the processing subject of various processes is as "program", in practice, it is needless to say that executes the processing CPU4 is based on the "program".
[0037]
　Figure 7 shows a processing procedure of the operation data acquisition process operation storage module 7 executes. Travel storage module 7, according to the processing routine shown in FIG. 6, to retrieve operational data from the elevator control device 11 of each of the elevators 3.
[0038]
　In practice, operation storage module 7, for example, initiate operation data acquisition processing shown in FIG. 7 in a time determined daily.
[0039]
　Then, operation storage module 7, first, from each of the elevators 3 of the elevator control device 11 obtains the operation data of one day (S11). Subsequently, operation storage module 7 stores the operational data for one day in the operation data table TB10 (S12), and ends the operation data acquisition process.
[0040]
　Figure 8 shows a processing procedure of the operation data learning processing which is executed by operating the learning module 8. Travel learning module 8, according to the processing routine shown in FIG. 8, on the basis of the operation data acquired from the operation data table TB10, a call button 16 at any respective hierarchical status and destination floor designated buttons for each elevator car 12 state for learns 5 minutes the state of destination floor specification button state and the elevator car 12 of the call button 16 in each layer of (1 time it takes to round trip) (learning result).
[0041]
　In practice, service learning module 8, for example, starts the operation data learning processing shown in FIG. 8 at a time that is determined every year.
[0042]
　Then, operation learning module 8 first acquires the operational data of one year from the operation data table TB10, learning based on the acquired operation data (S15). Subsequently, operating the learning module 8 stores the learning result as learning data in the learning data table TB20 (S16), and terminates the operational data learning processing.
[0043]
　Figure 9 shows a processing procedure of the service route determination process performed by the route determination module 9. Path decision module 9 according to the processing routine shown in FIG. 9 to determine the service route of the car 12.
[0044]
　In practice, the path decision module 9, the operation data learning processing is finished, to start the service route determination processing shown in FIG.
[0045]
　Then, the route determination module 9 first acquires learning data from the learning data table TB20 (S21). Subsequently, the path decision module 9, for each of the learning result, determines whether there is a path that can be omitted (S22). Path decision module 9 obtains a negative result in this determination, it sends to the normal path and the service route to the routing module 10, and ends the operation data learning processing.
[0046]
　In contrast, the path decision module 9, transmitted by that there is a path that can be omitted, if an affirmative result is obtained at the determination in step S22, generates a shortcut path is omitted the route (S23), the routing module 10 and, to end the operation data learning process.
[0047]
　Figure 10 illustrates a procedure of operation instruction process executed by the routing module 10. Routing module 10, according to the processing routine shown in FIG. 10, and instructs the service route to the elevator car 12.
[0048]
　In practice, the routing module 10, from the elevator control device 11 of the elevator 3, call upon receiving the passenger operations on buttons 16 and starts operation in the service route correction process shown in FIG. 10.
[0049]
　The routing module 10 first determines the elevator car 12 that runs (S25). Subsequently, routing module 10, the elevator control device 11 for controlling the elevator car 12 sends a service route (S26), and ends the operation instruction processing. The elevator controller 11 which has received the service route, the travel of the elevator car 12 in accordance with the service route.
[0050]
　Figure 11 illustrates a procedure of operation in the service route correction process executed by the routing module 10. Routing module 10, according to the processing routine shown in FIG. 11, to correct the service route of the car 12.
[0051]
　In practice, routing module 10, after the operation instruction processing is completed, the operation of the passenger against call button 16 of the landing operation in the path of the operation instruction processing car 12 instructed to travel by elevator controller 11 When receiving from starting the operation in the service route correction processing shown in FIG. 11.
[0052]
　The routing module 10 first determines whether or not there is elevator car 12 in close proximity to the landing from the respective elevator control device 11 acquires the position and the traveling direction of the car 12 (S31). Routing module 10, there are car 12 in close proximity to the landing, upon obtaining a positive result in this determination, it ends the operation in the service route correction process. Since the car 12 in this proximity to stop its landing, the control of the management server 2 is not required.
[0053]
　In contrast, routing module 10, by the absence of the car 12 in close proximity, upon obtaining a positive result in the determination at step S31, selects the closest car 12 to the landing (S32). Subsequently, routing module 10, the elevator control device 11 for controlling the elevator car 12 selected, transmits an instruction to reverse the traveling direction of the cab 12 (S33), and ends the operation in the service route correction process .
[0054]
　Figure 12 shows the processing procedure of the door opening when the service route correction process executed by the routing module 10. Routing module 10, according to the processing routine shown in FIG. 12, to correct the service route of the car 12.
[0055]
　In practice, routing module 10, after the operation instruction processing is completed, when the elevator car 12 which is instructed to travel by the operation instruction process is open the door (hereinafter, referred to as a door Hirakiji) upon receiving the pressing operation of the not yet occurred for the elevator car 12 (which should have occurred due to - forecasting of the learning result) button from the elevator control device 11, the door open during operation shown in FIG. 12 to start the path correction processing.
[0056]
　The routing module 10 first elapsed time from the expected time pressing operation of the button is occurring until this door open when it is determined whether more than a predetermined value (S41). Routing module 10, by a predetermined time has elapsed, it is determined that it is impossible to correct an error of - forecasting of the learning result, upon obtaining a negative result in this determination, the door open when the service route correction the process is terminated.
[0057]
　In contrast, routing module 10 is not predetermined time elapses, by determining that it is possible to correct an error of - forecasting of the learning result, a positive result in the determination of step S41 When an instruction to extend the time open the door, and transmitted to the elevator control device 11 controls the elevator car 12 (S42), and ends the time the door open service route correction process.
[0058]
(3) The implementation of effective embodiment
　the elevator management system 1 of the present embodiment as described above, the management server 2, predict whether the respective elevator 3 performs whatever operation the next period from the learning data by the instruction to perform the operation is omitted with respect to the path it can be omitted, performed for each of the elevators 3.
[0059]
　Therefore, according to the elevator management system 1, without modification of a program, it is possible to apply to the cage 12 take the operation according to the operating conditions, it can be operated efficiently cage 12.
[0060]
(4) Other Embodiments
　In the embodiment of yet above embodiment has dealt with the case of constituting the elevator management system 1 of the present invention as shown in FIG. 1, the present invention is not limited to this, these elevators the configuration of the management system, it is possible to widely apply various other configurations.
[0061]
　For example, as shown in FIG. 13, the elevator management system 20, the management server 2 and the respective elevator 3 may be configured such that connected via a communication network 21 such as the Internet. In this case, the management server 2 becomes cloud server, the server device installed in the data center. The management server 2, network 21 is connected to the elevator 3 through a communication path 24 such as a communication device 22, 23 and an intranet such as a switching hub or a router.
[0062]
　For the configuration as Figure 1, but hoistway and machine room of the elevator 3 it is assumed as the installation location of the management server 2, a large capacity such as can store operational data for one year in this place it may be difficult to install a server device so as to implement data storage and deep learning of. However, the elevator management system 20, with the construction shown in FIG. 13, it becomes possible to apply the present invention in such a case.
[0063]
　The EMS 20 is to have a connection to an external such as the Internet, for example, are installed in other buildings that are operated at the same working hours and hours, operational data of the elevator 3 to perform a similar operation it is possible to get. In this, the elevator management system 20, it is possible to obtain many operation data for learning, it is possible to improve the accuracy of the learning.
[0064]
　The EMS 20 has a connection to the external Internet, by information for learning the operation information of the weather data and public transport, it is possible to improve the accuracy of the learning.
[0065]
　Furthermore, as shown in FIG. 14, the EMS 30, the cloud server 35 for calculating the learning data, the communication network 21 and the management server 31 is a server device via the communication device 37, 39 such as a switching hub or a router and it may be connected to each of the elevator 3. Incidentally cloud server 35 is composed of a cloud server or data center or the like.
[0066]
　In the structure shown in FIG. 14, the process with a focus on operational data learning processing which requires transfer and learning process heavy operational data load is performed by high-performance cloud server 35, frequently communicate with the elevator 3 it is necessary to perform the processing by the communication delay around the operation instruction processing be fatal can be performed by the management server 31.
[0067]
　For this reason elevator management system 30, while reducing the influence of communication delay it can be installed in limited places relatively installation space. Incidentally, it is possible to perform rapid acquisition of the learning data by the learning data acquisition module 34 by installing a management server 31 to the DMZ (demilitarized zone), and operation instructions to the elevator 3.
[0068]
　Furthermore, as shown in FIG. 15, the communication of the elevator management system 50, by providing the communication module 54 to the management server 51 is a server device via the communication device 37 such as a switching hub or a router (Fig. 14) is not required. Therefore, it is also possible to apply the present invention when a switching hub or a router or the like for environmental reasons and security switching hub or a router or the like is not installed is not available. Incidentally, in the configuration of FIG. 15, only removes the management server 51 can be returned to operation of traditional elevators 3.
[0069]
　Furthermore, as shown in FIG. 16, when it is difficult to download the training data via the communication network, the EMS 60 may be connected to the auxiliary storage device 63 such as an SD card recording the learning data TB30. The auxiliary storage device 63, the maintenance personnel for the periodic maintenance and inspection of the elevator 3, updates the learning data TB30 when carrying maintenance and inspection work. Training data TB30 is created by copying the learning data TB20. Incidentally, in the configuration of FIG. 16, only removes the management server 61 is a server apparatus, can be returned to operation of traditional elevators 3.
[0070]
　Further, in the embodiment described above has dealt with the case of using deep learning as learning means, the present invention is not limited thereto, it may be used statistical techniques such as regression analysis, other than deep learning machine learning may be used.
[0071]
　Further, in the embodiment described above, the destination floor designated call button 16 and the elevator car 12 in each layer of one round trip after only pressing state of the destination floor designated button call button 16 and the elevator car 12 in each layer it has dealt with the case of predicting the pressing state of the button, the present invention is not limited to this, information of the season, such as the four seasons, and years and years of information such as the leap year that the Olympic Games will be held, time zone information, such as morning day and night or the like may be reflect.
[0072]
　Further, in the embodiment described above has dealt with the case without consideration of the local information in the building, the present invention is not limited to this, and get local information such as usage information of the conference room in the building from the channel 19 it may be reflected in the prediction results Te.
DESCRIPTION OF SYMBOLS
[0073]
　1,20,30,50,60 ...... EMS, 2,31,51,61 ...... management server, 3 ...... elevator, 4,32,52,62 ...... CPU, 5,63 ...... auxiliary storage device, 6,33,36,53,64 ...... memory, 7 ...... operation storage module, 8 ...... service learning module, 9 ...... path determination module, 10 ...... routing module, 11 ...... elevator control, 12 ...... cab, 13 ...... main rope 14 ...... counterweight, 15 ...... hoist 16 ...... call button, 19,24,38,40 ...... channel, 21 ...... communication network, 22,23,37,39 ...... communication equipment, 35 ...... cloud server.

WE CLAIM

A elevator management system for managing an elevator provided with a control device for operating over a car to a plurality of floors,
　comprising a management apparatus for managing the control device,
　the management device,
　destination floor specified information and, a receiving circuit for receiving the car call information,
　a memory for recording by accumulating received information of the receiver circuit,
　a controller based on the recorded information in the memory, to learn the operation tendency of the car,
　the management information and an output circuit for outputting to the control unit
　includes a,
　the controller,
　based on the result of the learning, destination floor designation information after the received information a predetermined time of the reception circuit, and predicts the car call information ,
　based on the result of the prediction after the predetermined time, the management information is formed so as to restrict the operation floor ranging from the car,
　the control device, based on the management information To control the operation of the car to have
　an elevator control system.
[Requested item 2]
　The control device,
　the car call to determine the information and floor to reach based on the prediction of the destination floor specified information, reverses the traveling direction of the car when it reaches to the floor
　elevator management according to claim 1, wherein system.
[Requested item 3]
　From the present time to the time it takes for a single operation of each of the car, to predict the car call information and the destination floor specified information for each of the car, to determine the travel route of each of the car
　elevator management system according to claim 1, wherein.
[Requested item 4]
　After the prediction of the car call information and the destination floor specified information, if the car call or destination floor specified other than predicted occurs,
　correct the service route based on the moving direction of the position and the car of the current of the cab to
　the elevator management system according to claim 1, wherein.
[Requested item 5]
　Based on the difference between the car call information and the destination floor or were based on the prediction of the specified information Please call and the actual car call and the time of occurrence of the destination floor designated as the time of occurrence of the destination floor specified, open the door of the car to adjust the time,
　the elevator management system of claim 1, wherein.
[Requested item 6]
　An elevator management system for managing an elevator,
　destination floor designation information of each of the car, to record the operation status of the car, including car call information or given from each of the floor, the first to learn the operation status and the server device,
　the second server apparatus to predict given the car call from each floor based on the learning
　with a
　travel of each car is controlled by a control device,
　said control device ,
　the second based on the car call predictions in the server device, the car call is to the exclusion of floors that are not expected to occur, determining a service route of each car
　EMS.
[Requested item 7]
　A elevator management method for managing a control system by a management device for operating over a car of an elevator into a plurality of floors,
　said management apparatus
　receives destination floor specifying information, and the car call information,
　receiving the received information is recorded cumulatively with,
　based on the received information, to learn the operation tendency of the car,
　it outputs the management information to the control device as a learning result,
　based on the learning result, the receiving information destination floor designation information after a predetermined time, and predicts the car call information,
　based on the result of the prediction to form the management information so as to limit the operation floor ranging from the car,
　the control device to, to control the operation of the car on the basis of the management information
　management method of the elevator.