CN111247078B

CN111247078B - Elevator analysis system and elevator analysis method

Info

Publication number: CN111247078B
Application number: CN201880068025.1A
Authority: CN
Inventors: 佐藤信夫; 浅原彰规; 星野孝道; 鸟谷部训; 羽鸟贵大; 吉川敏文; 北野佑; 下出直树
Original assignee: Hitachi Ltd
Current assignee: Hitachi Ltd
Priority date: 2017-10-30
Filing date: 2018-10-15
Publication date: 2022-05-03
Anticipated expiration: 2038-10-15
Also published as: JP7437353B2; JP2019081634A; JP6904883B2; JP2021143076A; CN114873395A; CN114873395B; CN111247078A; WO2019087760A1; JP2023086850A

Abstract

An elevator analysis system having a processor and a storage device connected to the processor, wherein the storage device stores the number of passengers who appear at an elevator-taking place on each floor of an elevator group to be controlled due to the use of an elevator, the processor predicts the number of passengers in the future based on the number of passengers stored in the storage device, determines an operation rule applicable to operation control of each car belonging to the elevator group and a control parameter set in each operation rule based on the predicted number of passengers in the future, and further outputs the determined operation rule and control parameter.

Description

Elevator analysis system and elevator analysis method

The priority of japanese patent application No.2017-209721, filed on 30/10/2017, and the entire contents of which are incorporated herein by reference.

Technical Field

The invention relates to a technique for analyzing group-controlled elevators.

Background

In large-scale buildings, a plurality of elevators are installed in parallel in order to improve the transportation capacity of the elevators, and a system for selecting an optimal car to serve when calling an elevator at a boarding place is introduced. Also, as the size of the building is enlarged and the number of elevators installed in parallel is increased, the group control apparatus can appropriately control the plurality of elevators to improve the quality of service such as reduction of waiting time of users. At this time, the group control device attempts the optimal control by predicting the use state of the elevator using the operation data and the like.

In patent document 1, the use demand is predicted and analyzed by using the feature amount of the ascending boarding ratio, the feature amount of the ascending descending, the feature amount of the descending boarding ratio, and the feature amount of the descending according to the number of passengers riding the car.

In patent document 2, a camera is installed in a hall, and the number of people taking a hall is counted. When the number of elevator waiting persons in a certain time is predicted, the average value of the waiting times in the same time in a specific period in the past is used as a predicted value.

In patent document 3, the number of passengers riding in a car is used. The current congestion state of a building and the past usage history are used to predict the congestion state.

In patent document 4, a camera is installed so as to take an image from the front of a waiting hall toward an entrance of a building, and a car is dispatched when it is detected that a person approaches the waiting hall.

Documents of the prior art

Patent document

Patent document 1: japanese laid-open patent publication No. 2014-172718

Patent document 2: japanese laid-open patent publication No. 2015-9909

Patent document 3: international publication No. 2017/006379

Patent document 4: japanese patent laid-open publication No. 2000-26034

Disclosure of Invention

Problems to be solved by the invention

Since the boarding time, boarding floors, alighting floors, and the number of boarding persons of the users located at the boarding locations are unknown, the control of the group control elevator is limited.

In

patent documents

1 and 3, since the number of passengers riding in the elevator car is used, the riding floor and the landing floor are known, but the riding place is not known. Therefore, it is difficult to perform control according to the change of the elevator riding place.

In

patent documents

2 and 4, the camera is attached to the boarding area, and therefore the boarding area is known, but the boarding floor and the alighting floor are not known. Therefore, it is difficult to perform control according to the situations of the boarding floor and the alighting floor. Furthermore, the cost of installing the camera is additionally calculated.

The present invention aims to perform control so that the user dissatisfaction is alleviated by shortening the waiting time of the user who is at the elevator taking place when the use time, the elevator taking floor, the elevator getting-off floor and the number of people who take the elevator are obtained from the data and the future congestion situation is further predicted.

Means for solving the problems

In order to solve at least one of the above problems, the present invention is an elevator analysis system including a processor and a storage device connected to the processor, characterized in that: the storage device is configured to store the number of passengers, which is the number of users who are present due to the use of an elevator, at an elevator boarding location on each floor of an elevator group to be controlled, and the processor predicts the number of future passengers based on the number of passengers stored in the storage device, determines an operation rule applicable to operation control of each car belonging to the elevator group and a control parameter set in each operation rule based on the predicted number of future passengers, and further outputs the determined operation rule and control parameter.

Effects of the invention

According to one aspect of the present invention, dissatisfaction with elevator-related users can be alleviated by implementing optimal elevator control. Problems, structures, and effects other than those described above will be apparent from the following description of the embodiments.

Drawings

Fig. 1A is a block diagram showing the overall configuration of a group control elevator control system according to an embodiment of the present invention.

Fig. 1B is a block diagram showing a hardware configuration of an analysis server of an embodiment of the present invention.

Fig. 2 is an explanatory diagram showing a relationship between processing and data in the group control elevator control system according to the embodiment of the present invention.

Fig. 3 is a sequence diagram showing an outline of the prediction of the number of passengers and the prediction of the destination floor in the process of the control system for the group control elevator according to the embodiment of the present invention.

Fig. 4 is a sequence diagram showing an overview of the effective rule/parameter selection in the process of the group control elevator control system according to the embodiment of the present invention.

Fig. 5 is a flowchart showing the processing of the occupant count estimation model generation unit according to the embodiment of the present invention.

Fig. 6 is a flowchart showing the processing of the occupant number estimation unit according to the embodiment of the present invention.

Fig. 7 is a flowchart showing the processing of the occupant number prediction unit according to the embodiment of the present invention.

Fig. 8 is a flowchart showing the processing of the destination floor estimation unit according to the embodiment of the present invention.

Fig. 9 is a flowchart showing the processing of the destination floor prediction unit according to the embodiment of the present invention.

Fig. 10 is a flowchart showing a process of controlling the selector section according to the embodiment of the present invention.

Fig. 11 is a flowchart showing the processing of the rule/parameter evaluation unit in the embodiment of the present invention.

Fig. 12 is an explanatory diagram showing building basic information stored in the analysis server according to the embodiment of the present invention.

Fig. 13 is an explanatory diagram showing a random seed held by the analysis server according to the embodiment of the present invention.

Fig. 14 is an explanatory view showing the number of persons who take in and out of the elevator stored in the analysis server according to the embodiment of the present invention.

Fig. 15 is an explanatory diagram showing an elevator operation log stored in the analysis server according to the embodiment of the present invention.

Fig. 16 is an explanatory diagram showing external information (weather) held by the analysis server according to the embodiment of the present invention.

Fig. 17 is an explanatory diagram showing external information (camera) held by the analysis server according to the embodiment of the present invention.

Fig. 18 is an explanatory diagram showing external information (building information) stored in the analysis server according to the embodiment of the present invention.

Fig. 19 is an explanatory diagram showing the occupant number estimation input stored in the analysis server according to the embodiment of the present invention.

Fig. 20 is an explanatory diagram showing the occupant count estimation model stored in the analysis server according to the embodiment of the present invention.

Fig. 21 is an explanatory diagram showing the result of estimating the number of occupants stored in the analysis server according to the embodiment of the present invention.

Fig. 22 is an explanatory diagram showing the result of prediction of the number of occupants stored in the analysis server according to the embodiment of the present invention.

Fig. 23 is an explanatory diagram showing the occupant number prediction result 2 stored in the analysis server according to the embodiment of the present invention.

Fig. 24 is an explanatory diagram showing destination floor estimates for different time periods stored by the analysis server according to the embodiment of the present invention.

Fig. 25 is an explanatory diagram showing destination floor prediction results for different time periods stored in the analysis server according to the embodiment of the present invention.

Fig. 26 is an explanatory diagram showing a rule/control template held by the analysis server according to the embodiment of the present invention.

Fig. 27 is an explanatory diagram showing a KPI list held by the analysis server according to the embodiment of the present invention.

Fig. 28 is an explanatory diagram showing simulation inputs and results held by the analysis server according to the embodiment of the present invention.

Fig. 29 is an explanatory diagram showing valid rules/parameters held by the analysis server according to the embodiment of the present invention.

Fig. 30 is an explanatory diagram showing an effective rule/parameter breakdown list held by the analysis server according to the embodiment of the present invention.

Fig. 31 is an explanatory diagram showing a rule/parameter list held by the analysis server according to the embodiment of the present invention.

Fig. 32 is an explanatory diagram showing a building personalized report output by the analysis server according to the embodiment of the present invention.

Detailed Description

Next, an embodiment of the present invention will be described in detail with reference to the drawings, but the present invention is not limited to the following embodiment, and various changes and modifications can be made within the technical concept of the present invention, and these changes and modifications are included in the scope of protection. Hereinafter, one embodiment of the present invention will be described with reference to fig. 1.

The analysis server SA, the client terminal CL, the external information neighborhood building information EXN, the external information database EXD, the external information camera EXC, the control cabinet CA, the car 1CA1, the car 2CA2, and the car 8CA8 are connected to the network NW in an open or close state.

The analysis server SA constitutes an elevator analysis system for performing an analysis relating to the control of group controlled elevators. The analysis server SA includes a database SA0, a display section SA1, a request section SA2, and an execution section SA 3.

The database SA0 handles input/output data used in the analysis server SA. Specifically, the database SA0 includes information preset in the analysis server SA, information acquired via the network NW, information generated by the processing of the execution unit SA3, and the like. Although not shown in fig. 1A, the database SA0 includes, for example, building basic information SA00, random seeds SA01, number of passengers on/off an elevator SA02, elevator operation logs SA03, outside information (weather) SA04, outside information (camera) SA05, outside information (building information) SA06, passenger number estimation inputs SA07, passenger number estimation models SA08, passenger number estimation results SA09, passenger number prediction results SA10, passenger number prediction results SA11, different-slot destination floor estimation SA12, different-slot destination floor prediction results SA13, rule/control template SA14, KPI list SA15, simulation inputs and results SA16, valid rule/parameter SA17, and rule/parameter list SA19 (see fig. 2, fig. 12 to fig. 31).

The execution unit SA3 is a unit that actually executes the analysis, and is configured by a measurement processing unit SA31, an occupant number estimation model generation unit SA32, an occupant number estimation unit SA33, an occupant number prediction unit SA34, a destination floor estimation unit SA35, a destination floor prediction unit SA36, a control selector unit SA37, and a rule/parameter evaluation unit SA 38.

The client terminal CL is a terminal used by the administrator to browse the analysis status. The external information neighborhood building information EXN, the external information database EXD, the external information camera ExC and the control cabinet CA provide information that is not relevant to the operation of the elevator. These pieces of information are described as external information. The external information may include common information such as railway traffic data and road condition data, in addition to the above information. The car 1CA1, the car 2CA2, and the car 8CA8 are elevator cars, and the control panel CA is a control device for controlling the cars 1CA1 to 8CA 8.

Although some of the cars are not shown in the figure, in the example of fig. 1A, 8 cars from car 1CA1 to car 8CA8 are controlled by the control cabinet CA. The cars 1CA1 to 8CA8 are cars of an elevator group to be group-controlled, which is formed of 8 elevators installed in the same building and facing the same elevator boarding area (waiting hall), for example. However, 8 elevators are only examples, and the present invention can also be applied to an elevator group including 2 or more elevators.

In the present embodiment, the number of passengers is the number of users who are present in the hall for riding the elevator.

Fig. 1B is a block diagram showing a hardware configuration of the analysis server SA of the embodiment of the present invention.

The analysis server SA is, for example, a computer having an interface (I/F)101, an input device 102, an output device 103, a processor 104, a main storage device 105, and an auxiliary storage device 106 connected to each other.

The interface 101 is connected to a network NW, and communicates with the client terminal CL, the external information database EXD, the external information camera EXC, and the control cabinet CA via the network NW, and obtains external information neighborhood building information EXN and the like. The input device 102 is a device used by a user of the analysis server SA to input information to the analysis server SA, and may include at least one of a keyboard, a mouse, and a touch sensor or the like. The output device 103 is a device for outputting information to the user of the analysis server SA, and may include a display device that displays, for example, characters and images.

The processor 104 executes various processes according to programs stored in the main storage 105. The main memory device 105 is a semiconductor memory device such as a DRAM, for example, and stores a program executed by the processor 104, data necessary for processing by the processor, and the like. The secondary storage device 106 is a storage device of a larger capacity, such as a hard disk drive or a flash memory, and stores data or the like referred to during processing performed by the processor.

The main storage 105 of the present embodiment stores programs for realizing the measurement processing unit SA31, the occupant number estimation model generation unit SA32, the occupant number estimation unit SA33, the occupant number prediction unit SA34, the destination floor estimation unit SA35, the destination floor prediction unit SA36, the control selector unit SA37, and the rule/parameter evaluation unit SA38 included in the execution unit SA 3. Therefore, in the following description, the processing executed by each part included in the execution unit SA3 is actually executed by the processor 104 in accordance with the program corresponding to each part stored in the main storage 105.

The processing of the request section SA2 may be realized by the processor 104 controlling the interface 101 or the input device 102 in accordance with a program corresponding to the request section SA2 stored in the main storage 105. The processing of the display section SA1 may also be realized by the processor 104 controlling the output device 103 according to a program corresponding to the display section SA1 stored in the main storage 105.

The auxiliary storage device 106 of the present embodiment stores a database SA 0. Further, the programs corresponding to the respective parts included in the execution part SA3 may be stored in the auxiliary storage device 106, or may be copied to the main storage device 105 as needed. In addition, at least a portion of database SA0 may also be copied to primary storage 105 as needed.

By referring to this figure, input data and output data relating to the respective processes are made more explicit. Further, an overall overhead of processing and data may be performed. The thick line boxes represent processing and the thin line boxes represent data. Further, the range enclosed by the solid line is preferably processed in real time, and the range enclosed by the broken line is preferably executed offline.

Specifically, the occupant number estimation model generation unit SA32 executes occupant number model processing SP01 (fig. 5) based on the building basic information SA00 (fig. 12) and the random seed SA01 (fig. 13), and outputs the occupant number estimation model SA08 (fig. 20). In the example of fig. 2, this occupant number model processing SP01 is performed as offline processing SZ 1.

The passenger number estimating unit SA33 executes the passenger number estimating process SP02 (fig. 6) based on the number of passengers on/off the elevator SA02 (fig. 14), the elevator operation log SA03 (fig. 15), the external information (weather) SA04 (fig. 16), the external information (camera) SA05 (fig. 17), the external information (building information) SA06 (fig. 18), the building basic information SA00, and the passenger number estimating model SA08, and inputs the result to the passenger number predicting unit SA 34. If there is available external information other than the above information, the occupant number estimation unit SA33 may use the external information.

The occupant number prediction unit SA34 executes the occupant number prediction processing SP03 (fig. 7) based on the result of the occupant number estimation processing SP02, and inputs the result to the destination floor prediction unit SA36 and the control selector unit SA 37. Further, this result is stored by the storing process SP07 (fig. 9).

The destination floor estimating unit SA35 performs destination floor estimating processing SP04 (fig. 8) based on the number of persons boarding and alighting the elevator SA02, and inputs the result to the destination floor predicting unit SA 36.

The destination floor prediction unit SA36 executes a destination floor prediction process SP05 (fig. 9) based on the results of the passenger number prediction process SP03 and the destination floor estimation process SP 04. This result is stored by the storing process SP 07.

The control selector SA37 executes the control selector SP06 (fig. 10) based on the result of the destination floor prediction processing SP05, the result of the passenger number prediction processing SP03, and the rule/parameter list SA19 generated by the display/control data generation processing SP15 described later, and outputs the result to the control panel CA.

In the example of fig. 2, the above described occupant number estimation processing SP02 to storage processing SP07 are executed as real-time processing SZ 0.

The rule/parameter evaluation unit SA38 executes KPI simulation processing SP11, effective rule/parameter selection SP12, end determination processing SP13, effective rule/parameter subdivision processing SP14, and display/control data generation processing SP15 (fig. 11) based on the data stored in the storage processing SP07, the rule/control template SA14 (fig. 26), and the KPI list SA15 (fig. 27). In the process, simulation input and results SA16 (fig. 28) and valid rules/parameters SA17 (fig. 29) are generated, and a rule/parameter list SA19 (fig. 31) and a building personalization report SA20 (fig. 32) are finally output.

In the example of fig. 2, the KPI simulation process SP11 to the display/control data generation process SP15 described above are executed as the offline process SZ 2.

The timing chart in fig. 3 is represented using 4 axes corresponding to each of the relevant data (the number of persons on/off the elevator SA02, the elevator operation log SA03, the external information (weather) SA04, the external information (camera) SA05, and the like), the analysis server SA, the control cabinet CA, and the client terminal CL.

The data collection S01 is a process of periodically transmitting data to the analysis server SA by an external system such as the external information database EXD. The measurement processing section SA31 of the execution section SA3 of the analysis server SA receives these data, performs database registration S02, and stores the received data in each table of the database SA 0. The processes of the occupant number estimation model generation unit SA32, the occupant number estimation unit SA33, the occupant number prediction unit SA34, the destination floor estimation unit SA35, the destination floor prediction unit SA36, and the control selector unit SA37 of the execution unit SA3 are periodically executed, and the result data is stored in each table of the database SA0 by database registration S03. Finally, the analysis server SA transmits the control parameters selected by the control selector section SA37 to the control cabinet CA as an input command CA 0.

The timing diagram of fig. 4 is represented using the same 4 axes as fig. 3.

In the client terminal CL, the manager performs the operator information input S04, such as inputting KPIs, time periods, and the like, as the operator information. The client terminal CL sends a request command including the entered information to the analysis server SA.

The analysis server SA executes data acquisition S05, and acquires data transmitted from the client terminal CL. Then, the analysis server SA executes the rule/parameter evaluation section SA38, selects useful rules and control parameters while acquiring corresponding data from the database SA0, and generates content using the result. The analysis server SA transmits display data including the generated content to the client terminal CL.

The client terminal CL executes the display processing S06, and displays the content. With regard to an example of the displayed content, it will be explained below with reference to fig. 30. Further, since KPIs, time periods, and the like are used in the analysis, it is preferable to perform recording in advance.

Fig. 5 is a flowchart showing the processing of the occupant number estimation model generation unit SA32 according to the embodiment of the present invention.

The occupant number model processing SP01 is composed of two parts, an occupant number data generation SP010 and an occupant number estimation model generation SP 011. In the passenger number data generation SP010, the passenger number estimation model generation unit SA32 obtains the number of passengers (that is, the number of passengers and the number of passengers getting off) and the car state of each car by simulation (2 nd simulation) based on the building basic information SA00 and the random seed SA 01.

For example, the occupancy estimation model generation unit SA32 randomly generates a plurality of users who will take an elevator in each hall during the simulation. Specifically, the occupant number estimation model generation unit SA32 randomly specifies the hall floor and the appearance time of each user using the random seed SA 01. Further, the occupant number estimation model generation unit SA32 randomly specifies the destination floor of each user from the floors selected from the building basic information SA 00.

Then, the number of passengers estimation model creation module SA32 simulates the operation of each car based on the specified appearance time, appearance floor, and destination floor of each user, determines the number of passengers getting on and off the car and the car state for each time, and creates this as the number of passengers estimation input SA 07. The car state is, for example, the floor where each car is located, the traveling direction (upward direction or downward direction) of each car, the number of passengers in each car, and the like, and may be the same as a value recorded in an elevator operation log SA03 described later. However, although the actual measurement values are recorded in the elevator operation log SA03, the occupant number estimation model generation section SA32 may generate the values by simulation.

At this time, the occupant number estimation model generation unit SA32 may execute the simulation based on any one of the operation rules and control parameters recorded in the rule/control template SA14 (fig. 26) described later, for example.

In the occupant number estimation model generation SP011, the occupant number estimation model generation unit SA32 generates the occupant number estimation model SA08 by executing the two-stage process composed of step 1 and step 2. In step 1, the occupant number estimation model generation unit SA32 specifies the number of occupants per time, the number of persons who take or land, and the car state, which are obtained in correspondence with each other, from the simulation result. Then, in step 2, the number-of-passengers estimation model generation unit SA32 obtains a function f satisfying a model for estimating the number of passengers, that is, the number of passengers satisfying f (the number of passengers getting on and off, the car state) from the state of each car, the number of passengers, and the number of passengers getting off and on each floor. As described later, for example, referring to fig. 20, the number of passengers and the state of the car may be used as explanatory indexes, and the number of passengers may be used as target variables to perform multiple regression analysis.

At this time, if at least one of the elevator operation log SA03, the external information (weather) SA04, the external information (camera) SA05, and the external information (building information) SA06 (or other external information) can be used, these values can be used as external variables to obtain a function f satisfying the number of passengers ═ f (the number of passengers, the state of the car, and the external variables). When the function f is obtained, it is only necessary to obtain the inverse conversion from the number of passengers determined in step 1 to the number of passengers on/off the elevator. Further, if the function f is found, other methods may be used.

Fig. 6 is a flowchart showing the processing of the occupant number estimation unit SA33 according to the embodiment of the present invention.

In the passenger number estimation processing SP02, the passenger number estimation unit SA33 substitutes the current number of passengers taking in and taking off the elevator SA02 and the actual number of passengers taking in and taking off the elevator SA08 obtained from the elevator operation log SA03 into the passenger number estimation model SA08 obtained in fig. 5 to obtain the current passenger number estimation result SA09, and stores it in the main storage device 105 or the auxiliary storage device 106. Thus, the riding condition of the user who is about to take the elevator can be estimated according to the riding condition of the user in each elevator car, the position, the traveling direction and other states of each elevator car.

The number of persons who get on and off the cars can be estimated from the change in weight of the cars measured by the control cabinet CA, for example. In addition, the position, the traveling direction, and the like of each car depend on the control of the control cabinet CA. Therefore, according to the above occupant number model processing SP01 and the occupant number estimation processing SP02, even when no external information is obtained, the riding situation of the user who is about to ride the elevator can be estimated based on the information obtained by the elevator itself.

In addition, the current number of persons boarding and alighting the elevator SA02 and the elevator operation log SA03 may also contain information for identifying the operation rules/control parameters applicable to the elevator when acquiring the data contained in these (that is, the operation rules/control parameters according to which the control cabinet CA performs the control of each car, see, for example, fig. 26). In this case, the occupant number estimation unit SA33 obtains the current occupant number estimation result SA09 by substituting the current elevator riding number SA02 and the actual elevator riding number, elevator descending number, and car state obtained from the elevator operation log SA03 into the occupant number estimation model SA08 generated by the occupant number estimation model generation unit SA32, based on the simulation of the operation rule and control parameter. Thereby, highly accurate estimation can be performed.

In addition, at this time, if at least any one of the elevator operation log SA03, the external information (weather) SA04, the external information (camera) SA05, and the external information (building information) SA06 (or other external information) can be used, these can be used in substitution.

Fig. 7 is a flowchart showing the processing of the occupant number prediction unit SA34 according to the embodiment of the present invention.

The occupant count prediction processing SP03 is constituted by an occupant count prediction SP030 and a format conversion SP 031.

In the occupant number prediction SP030, the occupant number prediction unit SA34 predicts the number of occupants in the future time from the time stored in the occupant number estimation result SA09 using the occupant number estimation result SA09 for each time point (for example, for each time slot having a predetermined time width) obtained in the processing of fig. 6, and outputs the result as the occupant number prediction result SA 10. At this time, if at least any one of the elevator operation log SA03, the external information (weather) SA04, the external information (camera) SA05, and the external information (building information) SA06 (or other external information) can be used, these can be used in substitution. For example, when the external information (camera) SA05 can be used, the number of persons (SA057) contained in the external information (camera) SA05 may be used instead of the number of persons who take the vehicle obtained from the number of persons estimation result SA 09; when the number of passengers determined from other external information can be used, these can also be used.

In the format conversion SP031, the occupant number prediction result SA10 obtained in the occupant number prediction SP030 is used to convert it into the probability of taking a person of a different number of persons per unit time using the poisson distribution, and the result is output as the occupant number prediction result 2_ SAll. This ride probability can be used to perform the later-described KPI simulation.

Fig. 8 is a flowchart showing the processing of the destination floor estimation unit SA35 according to the embodiment of the present invention.

The destination floor estimating unit SA35 estimates a destination floor by finding the tendency of people who get off the elevator car for each time slot based on the number of people who get on and off the elevator SA02, and outputs the result as destination floor estimation SA12 for different time slots.

Fig. 9 is a flowchart showing the processing of the destination floor prediction unit SA36 according to the embodiment of the present invention.

The destination floor prediction unit SA36 executes destination floor prediction processing SP05 and storage processing SP 07.

In the destination floor prediction processing SP05, the destination floor prediction unit SA36 predicts the destination floor of the seated user, that is, the floor to which the seated user will go. Specifically, the destination floor prediction unit SA36 predicts a destination floor by multiplying the passenger number prediction result 2_ SA11, which is the processing result of the passenger number prediction unit SA34 in fig. 7, by the different-slot destination floor estimation SA12, which is the processing result of the destination floor estimation unit SA35 in fig. 8, and outputs the result as the different-slot destination floor prediction result SA 13.

The storage processing SP07 is processing for storing the previous estimated number of passengers SA10 and destination floor estimation results SA13 for different time zones in, for example, a database SA 0. The reason for performing this processing is that a large amount of data in the past is required when performing offline processing.

Fig. 10 is a flowchart showing the process of controlling the selector section SA37 according to the embodiment of the present invention.

The control selector SP06 executed by the control selector section SA37 is a process for selecting a rule/parameter list that satisfies the occupant number prediction SA10 and the destination floor prediction SA13 at different time slots. The selected parameters are sent as input commands CA0 to the control cabinet CA.

Fig. 11 is a flowchart showing the processing of the rule/parameter evaluation unit SA38 according to the embodiment of the present invention.

The rule/parameter evaluation unit SA38 is configured from KPI simulation processing SP11, valid rule/parameter selection SP12, end determination processing SP13, valid rule/parameter subdivision processing SP14, and display/control data generation processing SP 15.

In the KPI simulation process SP11, the rule/parameter evaluation section SA38 executes a plurality of simulations (1 st simulation) while changing the operation rule/control parameters using the rule/control template SA14, the KPI list SA15, the different-period destination floor prediction result SA13, and the occupant number prediction result 2_ SA11, and outputs a KPI value.

Specifically, the rule/parameter evaluation unit SA38 presents users in the lobby of each floor according to the destination floor probability filled in the destination floor estimation SA12 and the riding probability for each number of passengers filled in the riding number prediction result 2_ SA11 for different time slots, and accordingly executes a simulation for controlling each car according to the operation rule/control parameter selected from the rule/control template SA 14. This simulation is performed multiple times while changing the applicable operating rules/control parameters, as described later.

In the valid rule/parameter selection SP12, the rule/parameter evaluation unit SA38 selects a valid rule/parameter from the values substituted into the KPI simulation processing SP11 and the results thereof.

In the end judgment processing SP13, the rule/parameter evaluation section SA38 judges whether or not the improvement effect can be seen as a result of the valid rule/parameter selection SP12, and if the improvement effect is seen, proceeds to "Yes"; if not, then "No" is entered.

In the effective rule/parameter subdivision processing SP14, the rule/parameter evaluation unit SA38 determines a range for subdividing more effective feature amounts based on the result of selecting SP12 from the effective rules/parameters.

The rule/parameter evaluation unit SA38 substitutes this result into the KPI simulation process SP11, and repeats the loop until the end determination process SP13 shows an improvement effect.

In the display/control data generation processing SP15, the rule/parameter evaluation section SA38 generates a rule/parameter list SA19 and a building personalized report SA20 based on the valid rule/parameter SA 17.

Fig. 12 is an explanatory diagram showing building basic information SA00 stored in the analysis server SA according to the embodiment of the present invention.

The building basic information SA00 is a table describing building basic information. The elevator of each building is composed of a plurality of cars, which is called an elevator group. The building basic information table is a table for managing each elevator group in order to perform control thereof (fig. 12). For example, car 1CA1 to car 8CA8 in fig. 1A belong to one elevator group. One elevator group corresponds to an elevator group as a group control object of the control cabinet CA. When there are multiple elevator groups in a building, there are many combinations of control cabinets and cars.

The building ID (SA000) is identification Information (ID) of the building in which the elevator is installed. Different IDs are used to identify the buildings. The elevator group ID (SA001) is an ID for distinguishing an elevator group in a building. The group name (SA002) is the name of the elevator group. The number of cars (SA003) is the number of cars constituting the elevator group. The target floor (SA004) indicates a floor at which a car constituting the elevator group stops. The latitude (SA005) and longitude (SA006) are respectively latitude and longitude indicating the location of the elevator group. The latitude and longitude of its center of gravity can also be used when the area of the elevator group is large. Further, it is sufficient if there is information indicating the position of the elevator group in absolute coordinates covering the entire earth, and may be values other than latitude and longitude. The building name (SA007) is the formal name of the building in which the elevator is located.

Fig. 12 shows an example, and if there is data necessary for the building basic information at the time of analysis, the building basic information SA00 may be changed and the data may be added.

Fig. 13 is an explanatory diagram showing a random seed SA01 held by the analysis server SA according to the embodiment of the present invention.

The random seed SA01 is a table describing seeds used when generating random numbers. The random seed No (SA010) is an ID of the random seed. A different ID is used to identify each random seed.

The random seed (SA011) is a random seed value. By using the present table, when a random seed No is specified, a value corresponding thereto can be referred to.

Fig. 13 shows an example in which, when generating a random number, if necessary data exists, the random seed SA01 may be changed to add the data.

Fig. 14 is an explanatory diagram showing the number of persons on/off the elevator SA02 stored in the analysis server SA according to the embodiment of the present invention.

The number of persons who get on/off the elevator SA02 is a table indicating the number of actual persons who get on/off the elevator in each car on each floor of the elevator.

The building ID (SA020) is an ID for identifying a building. The elevator group ID (SA021) is an ID for identifying each of a plurality of elevator groups in a building. Date (SA022) is a date for indicating the operation status of the elevator. Time (SA023) is time indicating the operation state of the elevator.

The week (SA024) is a week used to indicate the operation condition of the elevator. The time width (SA025) is a time width for summarizing the operation state of the elevator. Car 1(SA026) represents a car that has been identified as belonging to the elevator group identified by elevator group ID (SA 021). Floor (SA027) is the floor where car 1(SA026) was during the time period specified by date (SA022), time (SA023), week (SA024), and time amplitude (SA 025). The number of persons riding in the car (SA028) is the number of users riding the car (i.e., the number of persons in the car) in the time period specified by the date (SA022), time (SA023), week (SA024), and time width (SA 025).

The number of passengers (SA02A) riding in the upward direction (SA029) indicates the number of users (i.e., the number of passengers) riding the car when the car 1(SA026) is in the upward direction during a time period specified by the date (SA022), the time (SA023), the day of the week (SA024), and the time width (SA 025). The number of passengers getting off (SA02B) in the upward direction (SA029) indicates the number of passengers getting off (i.e., the number of passengers getting off) from the car when the car 1(SA026) is in the upward direction in the time period specified by the date (SA022), the time (SA023), the day of the week (SA024), and the time width (SA 025).

The number of passengers (SA02D) riding in the descending direction (SA02C) indicates the number of users riding in the car when the car is in the descending direction during a time period specified by the date (SA022), time (SA023), day of the week (SA024), and time width (SA 025). The number of passengers (SA02E) going down in the descending direction (SA02C) indicates the number of passengers when the car is in the descending direction during a time period specified by the date (SA022), the time (SA023), the day of the week (SA024), and the time width (SA 025).

The number of passengers on/off the elevator SA02 includes information on all the cars constituting the elevator group. One of the information related to the car 1(SA026) shown in fig. 14 is. Although not shown in fig. 14, data on other cars is also stored as the number of persons on/off the elevator SA 02.

The timing of filling the data in the elevator boarding/alighting number SA02 may be every event (for example, when a change actually occurs, etc.), or may be every predetermined period (for example, every 1 millisecond, every 1 second, every 1 minute, etc.). The date and time actually filled in can be represented by date (SA022), time (SA023), and week (SA 024). When information is filled in every predetermined period, the period may be filled as a time amplitude (SA 025). Further, all data specified in the present table need not be stored.

For example, the first row of the table in FIG. 14 represents: the car 1(SA026) in the elevator group identified by the elevator group ID "01" belonging to the building identified by the building ID "B001" was stopped at least once in 3 th within 5 minutes from 0 minute 1 second at 10 am on tuesday 10 am on 27 days 6 and 7 in 2017, the number of persons in the car (SA028) was 10, the number of persons who took the elevator (SA02A) and the number of persons who went the elevator (SA02bB) at the time of stopping during the upward movement were 15 persons and 1 person, respectively, and the number of persons who took the elevator (SA02D) and the number of persons who went the elevator (SA02E) at the time of stopping during the downward movement were 0 person and 10 persons, respectively. The number of people in the car (SA028) is the number of people who have completed boarding and alighting the elevator at the floor where the car is parked. If the car 1(SA026) stops several times on floor 3 within the above 5 minutes, these numbers may be the sum of these numbers, or may be filled in each stop on floor 3. In the same 5 minutes, when the car 1(SA026) stops at another floor more than once, the same information as described above is filled in the table of the floor.

Fig. 14 shows an example in which, when the number of persons who take the elevator at different floors is indicated, if necessary data is present, the number of persons who take the elevator SA02 may be changed to add the data.

Fig. 15 is an explanatory diagram showing an elevator operation log SA03 stored in the analysis server SA according to the embodiment of the present invention.

The elevator operation log SA03 is a table for indicating an actual elevator operation log. In this table, data can be stored both in summary in accordance with each elevator group and in car data belonging to the elevator group.

The building ID (SA030) is an ID for identifying a building. Elevator group ID (SA031) is an ID for identifying a plurality of elevator groups within a building. The date (SA032) is a date indicating the operation status of the elevator. Time (SA033) is time indicating the operation state of the elevator.

The week (SA034) is a week used to indicate the operation status of the elevator. The time width (SA035) is a width of time for summarizing the operation condition of the elevator. The long-time waiting rate (SA036) is a ratio of waiting time equal to or longer than a predetermined time length (for example, 60 seconds) in the waiting time generated in the elevator group (that is, the waiting time before the arrival of the car by the user calling the car) in the time period specified by the date (SA032), time (SA033), week (SA034) and time width (SA 035). The predetermined length of time may be altered by pre-specification.

The car call count (SA037) is the number of times the car call button is pressed within the elevator group in the time period specified by the date (SA032), time (SA033), week (SA034) and time width (SA 035). The traffic flow mode (SA038) is the operating mode of the elevator group.

The long time waiting rate (SA036), the number of car calls (SA037), and the traffic flow pattern (SA038) are values that are collected for each elevator group, but if necessary data is present, information other than the above may be added by changing the above.

Car 1(SA039) indicates that a car belonging to elevator group ID (SA031) has been identified. The floor (SA0A) is the position (floor) where the car 1(SA039) is located at the time point specified by the date (SA032), the time (SA033), and the week (SA 034). The direction (SA03B) is the travel direction of the car 1(SA039) at the time point specified by the date (SA032), time (SA033) and week (SA 034). For example, the upper table indicates an up direction, and the lower table indicates a down direction.

The state (SA03C) indicates the state of the car 1(SA039) at the time specified by the date (SA032), time (SA033), and day of the week (SA 034). For example, "action" indicates that the car 1(SA039) is actually running, and "stop" indicates that it has stopped. The number of passengers (SA03D) indicates the number of passengers riding in the car 1(SA039) at the time point specified by the date (SA032), time (SA033), and day of the week (SA 034).

The elevator operation log SA03 includes information relating to all the cars that constitute the elevator group. One of the cars 1(SA039) shown in fig. 15 is. Although not shown in fig. 15, data relating to other cars is also stored as an elevator operation log SA 03.

The timing of filling data in the elevator operation log SA03 may be every event (e.g., when a change actually occurs, etc.), or may be every predetermined period (e.g., every 1 millisecond, every 1 second, every 1 minute, etc.). The date and time actually filled in can be represented by date (SA032), time (SA033) and week (SA 034). When information is filled in every predetermined period, the period may also be filled as a time amplitude (SA 035). Further, all data specified in the present table need not be stored.

Fig. 15 shows an example in which, when an elevator operation log is displayed, if necessary data exists, the elevator operation log SA03 may be changed to add the data.

Fig. 16 is an explanatory diagram showing external information (weather) SA04 held by the analysis server SA according to the embodiment of the present invention.

The external information (weather) SA04 is a table for summarizing data related to weather as one of external information.

The external information ID (SA040) is an identification ID of the external information. The date (SA041) is the date when the external information was acquired. The time (SA042) is the time when the external information is acquired. The week (SA043) is the week in which the external information is acquired. The location (SA044) is a location where the external information is acquired. The latitude (SA045) is the latitude at which the extrinsic information is acquired. The longitude (SA046) is the longitude at which the external information is acquired. The weather (SA047), the air temperature (SA048), and the rainfall (SA049) are the weather, the air temperature, and the rainfall at the location specified by the location (SA044) at the time point specified by the date (SA041) and the time (SA042), respectively.

The timing of filling data in the external information (weather) SA04 may be every event (e.g., when a change actually occurs, etc.), or may be every predetermined period (e.g., every 1 millisecond, every 1 second, every 1 minute, etc.). The date and time actually filled in, and the place where the data was acquired may be represented by date (SA041), time (SA042), and place (SA 044). Further, all data specified in the present table need not be stored.

As an example, in the case of data indicating weather as one of the external information, if necessary data exists, the external information (weather) SA04 may be changed to add the data.

Fig. 17 is an explanatory diagram showing external information (camera) SA05 held by the analysis server SA according to the embodiment of the present invention.

The external information (camera) SA05 is a table for summarizing data relating to information identified by camera measurement, which is one of the external information.

The external information ID (SA050) is an identification ID of the external information. The date (SA051) is the date when this information was acquired. The time (SA052) is the time when the present information is acquired. The week (SA053) is the week on which this information was acquired. The building ID (SA054) is an ID for identifying the building for which the present information has been acquired. The floor (SA055) is the floor from which this information is obtained. The installation site (SA056) is a site where a camera is installed in order to acquire the present information.

The number of persons (SA057) is the number of persons detected by the camera installed at the point specified by the installation point (SA056) at the point in time specified by the date (SA051) and time (SA 052). The number of children (SA058), adults (SA059), men (SA05A), women (SA05B), wheelchairs (SA05C) and carts (SA05D) detected by the mounted cameras at the time point specified by the date (SA051) and time (SA052) and at the location specified by the mounting location (SA056), respectively, is the number of children, adults, men, women, wheelchairs and carts. In this way, not only the total number of users but also details of different user attributes (e.g., age group and gender), and objects other than the users can be detected.

From the results detected by the cameras installed at the points specified by the installation points (SA056) at the points in time specified by the date (SA051) and time (SA052), anger (SA05E) was the number of people who were judged to be angry users among the users detected by the cameras. Therefore, the number of the users can be detected, the emotion of the users can be detected through the facial expressions and behaviors by the aid of the camera, and the number of the users with the detected specific emotion can be further calculated.

The timing of filling data in the external information (camera) SA05 may be every event (e.g., when a change actually occurs, etc.), or may be every predetermined period (e.g., every 1 millisecond, every 1 second, every 1 minute, etc.). The date and time actually filled in, and the installation place of the camera acquiring this data may be represented by date (SA051), time (SA052), and installation place (SA 056). Further, all data specified in the present table need not be stored.

Fig. 17 shows an example in which, when data relating to information identified by camera measurement is indicated as one of the external information, if necessary data exists, the external information (camera) SA05 may be changed to add the data.

Fig. 18 is an explanatory diagram showing external information (building information) SA06 stored in the analysis server SA according to the embodiment of the present invention.

The external information (building information) SA06 is a table for summarizing data related to a building, which is one of the external information.

The external information ID (SA060) is an identification ID of the external information. The building ID (SA061) is an ID for identifying the building that has acquired the information. The date (SA062) is the date when the present information was acquired. Time (SA063) is the time to obtain this information. The week (SA064) is the week in which this information was acquired. The east side of the 3 th floor (SA065) indicates a floor (3 th floor) where the present information is acquired and an area (east side) where the present information is acquired in an area dividing the floor. The values summarized per floor and per area are stored. Floors and areas can be added arbitrarily, and when adding, data collected on the floors and areas can be stored, as in the east side of building 3 (SA 065).

The used amount (SA066) and the used amount (SA067) are the used amount and the used amount of water in the east side of the 3-storied building (SA065) at the time points specified by the date (SA062) and the time (SA063), respectively. Temperature (SA068) and humidity (SA069) are the temperature and humidity of the east side of the 3-storied building (SA065) at the time points specified by the date (SA062) and time (SA063), respectively. The number of stay (SA06A) is the number of stays on the east side of 3 stories (SA065) at the time point specified by the date (SA062) and time (SA 063).

The timing of filling data in the external information (building information) SA06 may be every event (e.g., when a change actually occurs, etc.), or may be every predetermined period (e.g., every 1 millisecond, every 1 second, every 1 minute, etc.). The date and time actually filled in, and the place where this data was acquired, can be represented by date (SA062), time (SA062), and 3-storied east side (SA 065). Further, all data specified in the present table need not be stored.

Fig. 18 shows an example in which, when data relating to a building, which is one of external information, is indicated, if necessary data exists, the external information (building information) SA06 may be changed to add the data.

Fig. 19 is an explanatory diagram showing the occupant number estimation input SA07 stored in the analysis server SA according to the embodiment of the present invention.

The occupant number estimation input SA07 is a table storing data generated by the occupant number model processing the occupant number data generation SP010 in SP 01. The generated data includes the number of passengers on each floor, the number of passengers getting on and off the different cars, and the car state.

The occupant number estimation input ID (SA070) is an ID for identifying the occupant number estimation input value. Time (SA071), week (SA072) and time amplitude (SA073) are respectively the time, week and time amplitude generated by the occupant number data generation SP 010. The number of passengers (SA074) is the number of passengers generated by the passenger number data generation SP 010. The number of passengers is obtained according to different floors. In fig. 19, the number of passengers in floor 3 is represented by floor 3 (SA 075). Although not shown in fig. 19, the number of passengers on the other floors is filled in the same manner. The number of passengers may be generated for each floor, each area, and each hall, and in this case, the generated number of passengers may be stored in the number of passengers (SA 074).

The number of persons who take the elevator in different cars (SA076) is the number of persons who take the elevator in different cars generated by the data generation SP010 of the number of persons who take the elevator in the time period specified by the time (SA071), the week (SA072) and the time width (SA 073). The number of persons who get on or off the elevator (SA076) for different cars is determined for each car. In fig. 19, the number of persons who get on and off the car 1 is indicated by the car 1(SA 077). Information on the number of passengers getting on and off the car 1 is stored in the car 1(SA077), the floor (SA078) is the floor where the car is located, the upward direction (SA079) is the number of passengers getting on and off the car when the upward running car stops at the floor, and the downward direction (SA07A) is the number of passengers getting on and off the car when the downward running car stops at the floor. In addition to the above, other information related to the car 1 can be stored in the car 1(SA 077). The number of persons who get on or off the car (SA076) is different from that of the car 1, and information on the car may be stored.

The car state (SA07B) stores data related to the car state, and stores data related to the car 1 in the car 1(SA 07C). The floor is the floor where car 1(SA07C) was located during the time period specified by time (SA071), week (SA072) and time amplitude (SA 073). The direction is the direction in which car 1(SA07C) travels during the time period specified by time (SA071), week (SA072), and time amplitude (SA 073). For example, "up" indicates an up direction and "down" indicates a down direction.

The state represents the state of the car 1(SA07C) in the time period specified by the time (SA071), the week (SA072) and the time amplitude (SA073) (SA 07C). For example, "action" indicates that it is actually running, while "docked" indicates that it is docked. The number of passengers indicates the number of passengers who take the car 1(SA07C) in the time period specified by the time (SA071), the week (SA072) and the time amplitude (SA 073). In addition to the above, information on the state of the car 1 may be stored in the car 1(SA 07C). Car state (SA07B) for cars other than car 1, information relating to the state of the car may be stored.

The timing of filling the data in the occupant number estimation input SA07 may be every event (e.g., when a change actually occurs, etc.), or may be every predetermined period (e.g., every 1 msec, every 1 sec, every 1 min, etc.). The date and time actually filled can be represented by time (SA071) and week (SA 072). Further, all data specified in the present table need not be stored.

Fig. 19 shows an example in which, when data generated in the occupant number data generation SP010 is shown, if necessary data exists, the occupant number estimation input SA07 may be changed to add the data.

Fig. 20 is an explanatory diagram showing the occupant number estimation model SA08 stored in the analysis server SA according to the embodiment of the present invention.

The occupant number estimation model SA08 is a table in which data generated by the occupant number estimation model generation SP011 is stored in the occupant number model processing SP 01. The generated data is a function f satisfying "number of passengers ═ f (number of passengers, car state, and external information)" when satisfied. The number of passengers, the number of persons who get on and off the elevator, and the state of the car are acquired as the estimated number of passengers input SA07, and the external information is acquired from external information (weather) SA04, external information (camera) SA05, and external information (building information) SA 06.

The occupant number estimation ID (SA080) is an ID for identifying an occupant number estimation model. The floor (SA081) is a floor serving as an object of the generated estimation model. The direction (SA082) is the direction of an object serving as the generated estimation model. Time (SA083) is the time of the object used as the generated estimation model. The week (SA084) is the week used as the object of the generated estimation model. The temporal amplitude (SA082) is the temporal amplitude of the object used as the generated estimation model.

The coefficients of the function f are stored in the back column. One or more items of data are selected as feature quantities from the number of passengers entering the different cars of the estimated number of passengers input SA07 (SA076), the car state (SA07B), the external information (weather) SA04, the external information (camera) SA05 and the external information (building information) SA 06. Then, these feature amounts are used as explanatory indexes and the number of passengers (SA074) is used as a target index, and coefficients of the feature amounts can be obtained by multiple regression analysis. The number of people on/off the elevator coefficient 1(SA085), the car state coefficient 1(SA087), and the external variable coefficient 1(SA088) are characteristic amount coefficients obtained by analysis. Since the coefficient for each feature amount is obtained, it is preferable that the coefficient for each feature amount is stored.

As a method for generating a model for estimating the number of passengers, an analysis method other than the multiple regression analysis may also be used.

The timing of filling the occupant number estimation model SA08 with data may be every event (e.g., when a change actually occurs, etc.), or may be every predetermined period (e.g., every 1 millisecond, every 1 second, every 1 minute, etc.). The date and time actually filled in can be represented by time (SA083) and week (SA 084). Further, all data specified in the present table need not be stored.

Fig. 20 shows an example in which, when representing a model generated in the occupant number estimation model generation SP011, if necessary data exists, the occupant number estimation model SA08 may be changed to add the data.

Fig. 21 is an explanatory diagram showing the occupant count estimation result SA09 stored in the analysis server SA according to the embodiment of the present invention.

The occupant number estimation result SA09 is a table in which data generated by the occupant number estimation SP020 is stored in the occupant number estimation processing SP 02. In the passenger number estimation SP020, the stored model (function f) of the number of passengers, the number of passengers boarding and alighting at the current time, the car state, and the external variables are inputted, and the number of passengers to be seated on each floor is estimated. This result is stored in the occupant number estimation result SA09 shown in fig. 21.

The occupant number estimation ID (SA090) is an ID for identifying the occupant number estimation. The date (SA092) is a date for estimating the number of passengers. The time (SA093) is a time for estimating the number of passengers. The week (SA094) is a week for estimating the number of passengers. The time width (SA092) is a time width for estimating the number of passengers. The floor (SA093) is a floor for estimating the number of passengers. The spot (SA094) is a spot for estimating the number of passengers. The number of passengers (SA095) is the number of passengers for estimating the number of passengers.

The timing of filling the data in the occupant number estimation result SA09 may be every event (for example, when a change actually occurs, etc.), or may be every predetermined period (for example, every 1 millisecond, every 1 second, every 1 minute, etc.). The date and time actually filled in can be represented by date (SA092), time (SA093), and week (SA 094). Further, all data specified in the present table need not be stored.

As an example, as shown in fig. 21, when the passenger number generated in the passenger number estimation SP020 is indicated, if necessary data exists, the passenger number estimation result SA09 may be changed to add the data.

Fig. 22 is an explanatory diagram showing the occupant number prediction result SA10 stored in the analysis server SA according to the embodiment of the present invention.

The occupant number prediction result SA10 is a table in which data generated by the occupant number estimation SP020 is stored in the occupant number prediction processing SP 03. In the occupant number prediction SP030, the occupant number prediction unit SA34 executes processing for estimating the number of future occupants using the occupant number estimation result SA09 obtained in the occupant number estimation processing SP02 and external information. This result is stored in the occupant number prediction result SA10 shown in fig. 22.

As explained below with reference to fig. 7, the input of the occupant number prediction SP030 includes a time width for analysis (for example, past 10 minutes), an occupant number estimation result SA09, outside information (weather) SA04, outside information (camera) SA05, and outside information (building information) SA06, and the output is a future occupant number.

As a method of predicting the number of passengers, for example, an AR model (autoregressive model) or the like may be used, but an analysis method other than the AR model may be used.

The occupant number prediction ID (SA100) is an ID that identifies that occupant number prediction has been performed. The date (SA101), time (SA102), and week (SA103) are the date, time, and week, respectively, which are the objects of analysis, i.e., at the time point at which the analysis was performed. The prediction time (SA104) is a time at which the analysis target is predicted (i.e., the number of passengers predicted at that time). The time width (SA105) is a time width of an analysis object. The floor (SA106) is a floor to be analyzed. The site (SA107) is a site of an analysis target. The number of passengers (SA108) is the number of passengers in the case of prediction analysis target.

For example, the first row of the result SA10 of the estimated number of passengers in fig. 22 shows that the process of estimating the number of passengers on the elevator floor of 3 floors is performed at 10 o ' clock 1 min/1 sec on the same day during 5 minutes from 6 o ' clock 2 nd tuesday 10 o ' clock 6 min 1 sec on 27 th day in 2017, and the result shows that the estimated number of passengers is 12.

The timing substituted in the occupant number prediction result SA10 may be every event (for example, when a change actually occurs, etc.), or may be every predetermined period (for example, every 1 millisecond, every 1 second, every 1 minute, etc.). The date and time actually filled in can be represented by date (SA101), time (SA102), and week (SA 103). Further, all data specified in the present table need not be stored.

As an example, as shown in fig. 22, when the occupant number prediction SP030 indicates the prediction of the number of occupants, if necessary data is present, the occupant number prediction result SA10 may be changed to add the data.

Fig. 23 is an explanatory diagram showing the occupant number prediction result 2_ SA11 stored in the analysis server SA according to the embodiment of the present invention.

The occupant number prediction result 2_ SA11 is a table for storing the result of format conversion of the occupant number prediction generated by the format conversion SP031 in the occupant number prediction processing SP 03.

In the format conversion SP031, the occupant number prediction unit SA34 obtains the probability of taking a person of different number of persons per unit time using the occupant number prediction result SA10 and the poisson distribution. This result is stored in the occupant number prediction result 2_ SA11 shown in fig. 23.

The formula of the poisson distribution is shown in the following formula (1). By substituting the number of passengers on different floors with λ in formula (1), the riding probability P (k) of k or more can be obtained.

[ number 1 ]

The above example is a method of determining the riding probability for each number of passengers assuming that the probability distribution of the number of passengers is a poisson distribution. However, in order to obtain the probability of the number of passengers, an analysis method other than the poisson distribution method may be used.

The occupant number prediction ID (SA110) is an ID for identifying that occupant number prediction has been performed. The date (SA111), time (SA112), and week (SA113) are the date, time, and week, respectively, which are the objects of analysis, i.e., at the time point at which the analysis was performed. The prediction time (SA114) is a time for which an analysis target is predicted (i.e., a ride probability at which the time is predicted). The time width (SA115) is a time width of an analysis object. The floor (SA116) is a floor to be analyzed. The site (SA117) is a site to be analyzed. The ride probability of more than one person (SA118) is a ride probability of more than one user per unit time. The ride probability of two or more persons (SA119) is a ride probability of two or more users per unit time. As the unit time, a time width (SA115) may also be used.

For example, the first row of the occupant number prediction result 2_ SA11 in fig. 23 shows an example corresponding to the prediction result filled in the first row of the occupant number prediction result SA10 in fig. 22. That is, the first row of the result of prediction of the number of passengers 2_ SA11 in fig. 23 indicates that, from the number of people "12 people" expected to take an elevator at the 3-th elevator floor, it is predicted that the probability of taking by one or more users at the 3-th elevator floor per unit time is 90% and the probability of taking by two or more users is 75%. Although not shown in fig. 23, in the same manner, the ride probabilities of three or more users and four or more users may be calculated and filled in the ride number prediction result 2_ SA 11.

As an example, as shown in fig. 23, when the format conversion SP031 indicates the prediction of the number of passengers, if necessary data is present, the number of passengers prediction result 2_ SAl1 may be changed to add the data.

Fig. 24 is an explanatory diagram showing destination floor estimates SA12 of different time slots held by the analysis server SA according to the embodiment of the present invention.

The different time slot destination floor estimation SA12 is a table storing data generated by the destination floor estimation process SP 04. In the destination floor estimation processing SP04, the destination floor estimation section SA35 generates a model for estimating the destination floors at different time slots using the number of boarding/alighting elevators at different floors (SA 02). Specifically, the destination floor estimation unit SA35 counts the number of persons going off at different floors for each time slot, and obtains the tendency of the number of persons going off at different floors. Then, it is converted into an estimated value of 100% as a whole. This result is stored in the different time period destination floor estimate SA12 shown in figure 24.

The destination floor estimation ID (SA120) is an ID for identifying the destination floor estimation that has been performed. The date (SA121), time (SA122), week (SA123), and time amplitude (SA124) are the date, time, week, and time amplitude of the analysis object, respectively. The boarding floor (SA125) is a boarding floor of a user. The direction (SA126) is the direction in which the car travels. The destination floor (SA127) is a user's landing floor. The estimate of 100% is filled in as a whole with respect to the floor at which the elevator stops.

For example, the first row in fig. 24 shows that 10% of users who ride the car in the upward direction on the 3 rd floor are getting off at the 26 th floor and 10% are getting off at the 27 th floor within 60 minutes from 1 minute 1 second at 10 am on tuesday on 27 th day in 2017, which is estimated from the number of people who get on and off the elevator SA 02. Although the proportion of users who get on the elevator at other floors is omitted in fig. 24, the sum of the proportions of all the destination floors of users who get on the elevator at floor 3 is calculated to be 100%. The same is true for the destination floors of users boarding other floors. In the present embodiment, these ratios can be used as destination floor probabilities, i.e., the probabilities that the destination floor of a user who appears at the boarding location of each floor is the floor.

For example, when it is possible to determine whether or not the user who takes the elevator car at each floor and the user who takes the elevator car from the car at each floor are the same person from the external information (camera) SA05 or the like, it is determined which floor the user who takes the elevator car at each floor is going to take, based on the determination result, and the destination floor ratio of the user who takes the elevator at each floor is calculated based on the determination result, and for example, the ratio of the users who take the elevator at the 26 th floor among the users who take the elevator at the 3 rd floor is 10%. However, when the external information cannot be used, for example, when it is estimated from the car weight that each of the boarding/alighting users cannot be identified such as the number of boarding/alighting persons at each floor, the destination floor ratio can be approximately calculated based on some assumption.

For example, the number of users who get off the elevator at each floor may be counted up in a time period specified by the date (SA121), the time (SA122), the day of the week (SA123), and the time width (SA124), and the ratio of the number of users who get off the elevator at the 26 th floor to the number of users who get off the elevator at floors other than the 3 rd floor may be calculated as the ratio of users who get off the elevator at the 26 th floor to users who get on the elevator at the 3 rd floor (i.e., the probability that the destination floor of the users who get off the elevator at the 3 rd floor is the 26 th floor). In this case, the same method can be used to calculate the proportion of users who get off the elevator at other floors and the proportion of users who get off the elevator at each floor after taking the elevator from other floors.

As an example, as shown in fig. 24, when the destination floor estimation is indicated in the destination floor estimation processing SP04, if necessary data exists, the destination floor estimation SA12 may be changed to add the data in a different time zone.

Fig. 25 is an explanatory diagram showing destination floor prediction results SA13 stored in the analysis server SA in different time zones according to the embodiment of the present invention.

The different-slot destination floor prediction result SA13 is a table for storing data generated by the different-slot destination floor prediction SP051 in the destination floor prediction processing SP 05. In the different-slot destination floor prediction SP051, the destination floor prediction section SA36 uses the different-slot destination floor estimation SA12 and the passenger number prediction result 2_ SA11 as input data, and by combining these, it is possible to predict to which floor the boarding user is going. Specifically, the probability of taking a call for predicting the boarding time may be multiplied by the destination floor estimate at the same time for different floors.

The prediction method of the destination floors of different time periods as shown above is an example, and other methods may be employed. This result is stored in the different time period destination floor prediction result SA13 shown in fig. 25.

The destination floor prediction ID (SA130) is an ID for identifying that destination floor prediction has been performed. The date (SA131), time (SA132), and week (SAl33) are the date, time, and week, respectively, which are the objects of analysis, i.e., at the time point at which the analysis was performed. The prediction time (SA134) is a time for which an analysis target is predicted (i.e., a ride probability at which the time is predicted). The time amplitude (SA135) is a time amplitude of an analysis object. The elevator boarding floor (SA136) is an analysis target elevator boarding floor. The destination floor (SA137) is a destination floor to be analyzed. The direction (SA138) is the running direction of the car to be analyzed. The riding probability of one or more persons (SA139) is the riding probability of one or more persons per unit time. The riding probability of two or more persons (SA13A) is the riding probability of two or more users per unit time. As the unit time, a time width (SA135) may also be used.

For example, the first row of the different time slot destination floor prediction SA13 in fig. 25 shows an example where the prediction filled in the first row of the passenger number prediction 2_ SA11 in fig. 23 corresponds to the estimation filled in the first row of the different time slot destination floor estimation SA12 in fig. 24. That is, the first row of the destination floor prediction results SA13 in different time zones in fig. 25 shows that, among users who plan elevators on the 26 th floor with cars traveling in the upward direction on the 3 rd elevator floor in a unit time, the riding probability of one or more persons is expected to be 9%, and the riding probability of two or more persons is expected to be 7.5%.

In this example, "9%" is obtained by multiplying a proportional value "10%" of the 26 th floor to the destination floor (SA127) in the first row in fig. 24 by "90%" of the riding probability (SA118) of one or more persons in the first row in fig. 23. "7.5%" is obtained by multiplying a proportional value "10%" of the 26 th floor to the destination floor (SA127) in the first row in fig. 24 by "75%" which is an occupancy probability (SA119) of two or more persons in the first row in fig. 23.

As an example, as shown in fig. 25, when different-time-zone destination floor predictions SP051 indicate different-time-zone destination floor predictions, if necessary data exists, the different-time-zone destination floor prediction result SA13 may be changed to add the data.

Fig. 26 is an explanatory diagram showing a rule/control template SA14 held by the analysis server SA according to the embodiment of the present invention.

The rules/control template SA14 is a table for storing operation rules/control parameter templates of elevators. Here, the operation rules refer to rules that the control panel CA applies to control the operation of a plurality of cars of an elevator to be group-controlled, and the control parameters refer to parameters that can be changed in each operation rule. In the present embodiment, the operation rule and the control parameter included therein are described as the operation rule/control parameter. In addition, the operation rule may be described as a rule alone, and the control parameter may be described as a parameter alone.

By using the rules/control template SA14, the best operating rules/control parameters may be searched. The search method consists of 2 steps. Step 1 is a search rule/control No (SA 140). This is the step of selecting a control parameter from a plurality of operating rules/control parameters that is suitable for improving the KPI. Step 2 is to search for a parameter value (initial value) (SA 144). The search object is a controllable parameter value among the control parameters. By searching for this, better control parameters can be found.

The rule/control No (SA140) is an ID for identifying an operation rule/control parameter. The rule name (SA141) is a name of the operation rule/control parameter. The condition (SA142) is an action condition of the operation rule/control parameter. The parameter value (initial value) (SA143) is a controllable parameter among the operation rules/control parameters. For example, in the rule "5 minutes after" corresponding to rule/control No "Ru 01", going from o layer to "the o portion (in this example, the number of floors) is a controllable parameter. The coefficients (initial coefficients) (SA145) are coefficients for obtaining a regression equation or the like. The parameter value (initial value) (SA143) and the coefficient (initial coefficient) (SA145) can change the stored values by repeatedly performing the optimization process.

Fig. 26 shows an example in which, when the operation rule/control parameter of the elevator is realized, if necessary data exists, the rule/control template SA14 may be changed to add the data.

Fig. 27 is an explanatory diagram showing a KPI list SA15 held by the analysis server SA according to the embodiment of the present invention.

The KPI list SA15 is a table storing KPIs (key performance indicators) as evaluation indicators when searching for optimal operation rules/control parameters. Since the KPI may differ from building to building, the KPI for each building is set in advance by using a flag (SA 155). At this time, a target value of KPI may also be set (SA 154).

KPIID (SA150) is an ID for identifying KPI. The classification (SA151) is a classification of KPIs. In particular, classification (SA151) indicates that criteria would benefit from improving this KPI.

The name (SA152) is the name of the KPI. The condition (SA153) indicates the content of the KPI. The target value (SA154) represents a target value of a variable parameter value portion (portion of O in the example of fig. 27) under the condition (SA 153). The setting is performed before use, because the situation is different for each building. KPIs used when performing the current optimization are specified from among the KIPs using flags (SA 155). When the use flag (SA155) is 1, it indicates that it has been designated. In addition, multiple KPIs may also be specified.

In the example of fig. 27, as KPIs, waiting time before taking a car of a user, a congestion rate at a boarding place, and a used amount of electricity (i.e., an amount of electricity including car travel) of a floor, which appear at the boarding place, are shown. In these examples, the following may be evaluated as appropriate operating rules/control parameters: for example, the operation rule/control parameter that can shorten the longest waiting time, the operation rule/control parameter that can reduce the congestion rate at the boarding place, and the operation rule/control parameter that can reduce the amount of electricity used.

However, the above is merely an example, and KPIs other than the above may be specified. For example, a KPI that is evaluated to be higher when the rate at which a plurality of users who take elevators on different floors take an elevator on the same car is smaller may be used. Thereby, control of the car can be achieved according to the requirements of the elevator-related user (e.g. user or supervisor, etc.) so that the user is satisfied.

Fig. 27 shows an example in which when the operation rule and control parameter of the elevator are realized, if necessary data exists, the KPI list SA15 may be changed to add the data.

Fig. 28 is an explanatory diagram showing the simulation input and result SA16 held by the analysis server SA according to the embodiment of the present invention.

The simulation input and result SA16 is a table for storing the result of processing performed by the KPI simulation process SP 11. In the KPI simulation process SP11, as inputs, a KPI list SA15 is used in which the occupant number prediction result 2_ SA11 indicating the riding situation, the different-period destination floor prediction result SA13, the rule/control template SA14 indicating the control parameters, and KPI as the optimization target are stored. By using these data, it is possible to find operation rules/control parameters for improving the KPI of the user in the boarding state.

In the KPI simulation processing SP11, while the operation rule/control parameter is changed, the KPI output processing when the user is in the elevator-riding state and uses a certain operation rule/control parameter is executed a plurality of times. The result is a simulated input and result SA 16.

The KPI simulation ID (SA160) is an ID that identifies a KPI simulation. The number of times (SA161) is the number of times when KPI simulation is performed a plurality of times. Rule control list 1(SA162) represents one set of operation rules/control parameters used in each simulation. The rule/control No (SA163) is an ID for identifying an operation rule/control parameter. The parameter value (SA164) is a control parameter for the current control. The coefficient (SA165) is a coefficient for determining a regression equation or the like. In one simulation, multiple rule control lists may be stored. KPIID (SA166) is an ID for identifying KPI. The KPI simulation result (SA167) is a KPI value obtained as a result when KPI simulation is performed using the rule control list.

Fig. 28 shows an example in which, when the operation rule and control parameter of the elevator are realized, if necessary data exists, the simulation input and result SA16 may be changed to add the data.

Fig. 29 is an explanatory diagram showing validity rules/parameters SA17 held by the analysis server SA according to the embodiment of the present invention.

The validation rules/parameters SA17 are tables for storing results of finding the operating rules/control parameters that contribute to optimization (that is, validation) from the simulation inputs and results SA16 shown in fig. 28. The rule/parameter evaluation unit SA38 is capable of performing multiple regression analysis by using the simulation input and result SA16 shown in fig. 28 as inputs, the target variables as KPI simulation results, and the explanatory variables as rule control lists. However, as long as the rule control parameters contributing to the optimization are specified, a method other than the multiple regression analysis method may be employed.

The valid rule/parameter ID (SA170) is an ID for identifying a valid operation rule/control parameter. The effective rule control list 1(SA171) is a rule control parameter that contributes most when the multiple regression analysis is performed. The rule/control No (SA172) is an ID identifying an operation rule/control parameter. The parameter value (SA173) is a control parameter value used in the present process. The coefficient (SA174) is a coefficient obtained by multiple regression analysis, and is a value indicating a degree of contribution to optimization. By reference to this, it is possible to specify effective (that is to say contributing to improving KPIs) operating rules/control parameters. A plurality of valid rule control lists may be stored. KPIID (SA175) is an ID for identifying KPI. The predicted value (SA176) is a KPI value predicted using a regression equation obtained by multiple regression analysis.

Fig. 29 shows an example in which, when the operation rule/control parameter of the elevator is realized, if necessary data exists, the validity rule/parameter SA17 may be changed to add the data.

Fig. 30 is an explanatory diagram showing an effective rule/parameter breakdown list SA18 held by the analysis server SA according to the embodiment of the present invention.

Further optimization can be achieved by subdividing the control parameter values for the operating rules/control parameters that are highly conducive to optimization as specified in the validation rules/parameters SA17 shown in fig. 29. The operation rule/control parameter having the larger coefficient (SA174) is selected in the active rule control list of the active rule/parameter SA 17. Then, the rule/parameter evaluation unit SA38 performs a process SP14 of subdividing the selected operation rule/control parameter into valid rules/parameters. Specifically, the rule/parameter evaluation unit SA38 may search for a more optimal operation rule/control parameter by increasing or decreasing the control parameter value included in the selected operation rule/control parameter.

The valid rule/parameter refinement ID (SA180) is an ID for identifying valid rule/parameter refinement. The valid rule/parameter ID (SA181) is an ID for identifying a valid operation rule/control parameter. The effective rule control list 1(SA182) is a rule control parameter that is estimated to contribute most by multivariate regression analysis. The rule/control No (SA183) is an ID identifying the operation rule/control parameter. The parameter value (SA184) is a control parameter value used in the present process. The coefficient (SA185) is a coefficient obtained by multiple regression analysis, and is a value contributing to optimization. The parameter value subdivision range (SA186) is a value obtained by the validation rule/parameter subdivision processing SP 14. Multiple active rule control lists may be stored. KPIID (SA187) is an ID for identifying KPI. The predicted value (SA188) is a KPI value predicted using a regression equation obtained by multiple regression analysis. Further, the rule/parameter evaluation section SA38 may randomly select several operation rules/control parameters from the rule/control template SA 14.

Fig. 30 shows an example in which, when the operation rule and control parameter of the elevator are realized, if necessary data exists, the valid rule/parameter breakdown list SA18 may be changed to add the data.

Fig. 31 is an explanatory diagram showing a rule/parameter list SA19 held by the analysis server SA according to the embodiment of the present invention.

The rule/parameter list SA19 is a table storing operation rules/control parameters used in selecting actual operations from the valid rules/parameters SA17 of fig. 29. Of the operation rules/control parameters in the valid rule control list, the portion having a large value of the determination coefficient (SA174) is the operation rule/control parameter having a high contribution rate.

The rule/parameter ID (SA190) is an ID that identifies an operation rule/control parameter. The first effective rule control (SA191) estimates a rule control parameter having the largest contribution degree from the result of the multiple regression analysis. The rule/control No (SA192) is an ID identifying an operation rule/control parameter. The parameter value (SA193) is a control parameter value used in the present process. The coefficient (SA194) is a coefficient obtained by multiple regression analysis, and is a value contributing to optimization.

The second active rule control (SA195) estimates the operation rule/control parameter of the second contribution degree bit sequence based on the result of the multiple regression analysis. The rule/control No (SA196) is a control parameter value used in the present process. The parameter value (SA197) is a control parameter value used in the present process. The coefficient (SA198) is a coefficient obtained by multiple regression analysis, and is a value contributing to optimization.

KPIID (SA199) is an ID for identifying KPIs. The predicted value (SA19A) is a value predicted using a regression equation obtained by multiple regression analysis.

The rule/parameter list SA19 is sent to the control selector SP 06. The control selector SP06 generates an input command CA0 indicating the operation rules/control parameters for improving the KPI according to the rule/parameter list SA19, and sends it to the control cabinet CA. The control cabinet CA changes the running rule/control parameter that has been set to the instructed running rule/control parameter according to the input command CA0, and controls the car according to the changed running rule/control parameter. Therefore, the control of the elevator after the KPI is improved is realized.

Fig. 31 shows an example in which, when the operation rule/control parameter of the elevator is realized, if necessary data exists, the rule/control parameter list SA19 may be changed to add the data.

The processing described in the present embodiment is executed in the execution unit SA3 of the analysis server SA, but a part or all of the processing may be executed by the control cabinet CA. For example, the control cabinet CA may have hardware similar to the analysis server SA shown in fig. 1B, and at least a part of the functions of the analysis server SA may be implemented by these hardware.

Fig. 32 is an explanatory diagram showing a building personalized report SA20 output by the analysis server SA according to the embodiment of the present invention.

The building individualization report SA20 is generated by the rule/parameter evaluation unit SA38 in the display/control data generation process SP15, and is sent to the display unit SA 1. The display unit SA1 (e.g., a display device installed as the output device 103) displays the received building personalized report SA 20.

For example, as shown in fig. 32, the building personalized report SA20 includes a building name 3201, an elevator group name 3202, a time period 3203, a KPI3204, and a result 3205.

The elevator group name 3202 and the building name 3201 are elevator group and building names of the elevator group installed as execution targets of various processes shown in fig. 2, and correspond to the group name (SA002) and the building name (SA007) shown in fig. 12. The period 3203 is a period as a simulation object. The KPI3204 is an evaluation index selected as an evaluation target in the processing of the rule/parameter evaluation section SA38, and corresponds to a KPI in which the use flag (SA155) shown in fig. 27 is valid. The result 3205 is a valid operation rule/control parameter selected according to the processing result of the rule/parameter evaluation section SA38, and corresponds to the operation rule/control parameter recorded in the rule/parameter list SA 19.

By referring to the building personalization report SA20, the elevator manager can grasp the contents of the changes in the operation rules/control parameters, which are required to improve the evaluation index displayed as the KPI 3204. The administrator can manually set the change of the grasped operation rule/control parameter in the control cabinet CA. Therefore, the control of the elevator after the KPI is improved is realized.

As described above, according to the present embodiment, the number of passengers in the hall is predicted from the number of elevator passengers on the floor, a control method suitable for the result of the prediction is generated, and an optimal elevator control can be realized by performing evaluation using an index relating to user dissatisfaction. For example, near the time when congestion is expected to occur in the future, by smoothly allocating cars at the elevator-taking place, it is possible to reduce the situation where the user waits for a long time at the elevator-taking place, improve the user's transport capacity, and further improve the user satisfaction.

The present invention is not limited to the above-described embodiments, but includes various modifications. For example, the above-described embodiments have been described in detail to better understand the present invention, but the present invention is not limited to having all the structures described.

Further, the above-described respective structures, functions, processing units, processing devices, and the like may be designed, for example, by integrated circuits, and a part or all of them may be realized by hardware. The respective structures, functions, and the like described above can be realized by software by interpreting and executing programs for realizing the respective functions by a processor. Information such as programs, tables, and files for realizing various functions may be stored in a storage device such as a nonvolatile semiconductor memory, a hard disk Drive, and a SSD (Solid State Drive), or may be stored in a non-transitory data storage medium readable by a computer such as an IC card, an SD card, and a DVD.

Furthermore, the control lines and information lines are shown for better illustration, but not necessarily all of them on the product. In practice, it can also be said that almost all structures are connected to each other.

Claims

1. An elevator analysis system, which is an elevator analysis system having a processor and a storage device connected to the processor,

the method is characterized in that:

the storage device stores: elevator taking in and out number information indicating the actual number of elevator taking in and taking out of each elevator car belonging to the elevator group to be controlled on each floor; running log information indicating an actual state of each car belonging to the elevator group; and the number of passengers, wherein the number of passengers is the number of users who appear at the elevator taking position of each floor of the elevator group due to the use of the elevator,

the processor performs the following operations:

estimating the number of passengers according to the number information of the passengers on the elevator and the running log information,

predicting a future number of passengers based on the estimated number of passengers,

and determines an operation rule applicable to operation control of each of the cars belonging to the elevator group and a control parameter set in each operation rule based on the predicted number of future passengers,

and further outputting the determined operation rule and the control parameter.

2. The elevator analysis system according to claim 1,

the method is characterized in that:

the processor calculates the probability of a destination floor according to the number of people taking the elevator, wherein the probability of the destination floor is the probability that the destination floor of the user appearing at each floor taking the elevator is the floor;

and determining an operation rule applicable to operation control of each car belonging to the elevator group and a control parameter set in each operation rule according to the number of future passengers and the destination floor probability.

3. The elevator analysis system according to claim 2,

the method is characterized in that:

the storage device further stores information specifying an evaluation index for evaluating the operation state of each car in the elevator group;

the processor

Executing a first simulation a plurality of times while changing the applicable operation rule and the control parameter, wherein the simulation generates users at elevator taking positions of each floor and then operates each elevator car in the elevator group according to the number of future passengers and the destination floor probability;

calculating the specified evaluation index based on the result of the first simulation,

the operation rules and the control parameters contributing to the improvement of the evaluation index are determined as the operation rules applicable to the operation control of each of the cars belonging to the elevator group and the control parameters set in the respective operation rules, based on the calculated evaluation index.

4. The elevator analysis system according to claim 3,

the method is characterized in that:

the processor calculates the riding probability of the users of the number of people according to each riding number when the probability distribution of the riding numbers is assumed to be Poisson distribution according to the predicted future riding number;

the user is generated according to the probability of taking for each number of passengers and the probability of destination floor for each destination floor, and the first simulation is performed.

5. The elevator analysis system according to claim 3,

the method is characterized in that: further having a display device coupled to the processor,

the processor specifies the operation rule and the control parameter that contribute to increase in the contribution amount of the evaluation index so as to satisfy a predetermined condition,

the display device displays the specified operation rule and control parameter.

6. The elevator analysis system according to claim 3,

the method is characterized in that:

further having the processor and an interface connected to an external network of the elevator analysis system;

a control device is connected to the network, wherein the control device is used for controlling each car belonging to the elevator group;

the processor specifies the operation rule and the control parameter which contribute to improving the contribution degree of the evaluation index and meet a preset condition;

sending the specified operation rule and control parameters to the control device via the interface.

7. The elevator analysis system of claim 3, wherein:

the evaluation index includes any one of a waiting time before a user who takes the elevator rides on any one of the cars, a congestion rate at the elevator riding place, and power consumption for operating each car in the elevator group.

8. The elevator analysis system according to claim 1,

the method is characterized in that:

the processor generates a plurality of users who appear at the elevator taking place of the elevator group due to the use of the elevator, randomly determines the elevator taking place floor where each user is located, the appearance time of each user, and the destination floor of each user, executes a second simulation for operating each car belonging to the elevator group according to the appearance time of each user, the elevator taking place floor where each user appears, and the destination floor, thereby generating a passenger number estimation model for estimating the number of passengers who appear at the elevator taking place of each floor, based on the state of each car, the number of passengers who take each car in each floor, and the number of passengers who take each car in each floor,

estimating the number of passengers on each floor by applying the actual number of passengers, the actual number of passengers getting on and off, and the state of each cage, which are obtained from the information on the number of passengers getting on and off the elevator and the operation log information, to the estimated number of passengers,

saving the estimated number of rides in the storage device,

predicting the future number of rides based on the estimated number of rides saved in the storage device.

9. The elevator analysis system of claim 8, wherein:

in the second simulation, the processor generates the estimated number of passengers by performing multiple regression analysis using the number of passengers on each floor per predetermined time width as a target index, and using the state of each car, the number of passengers on each car on each floor, and the number of passengers on each car on each floor as explanatory indices.

10. The elevator analysis system of claim 8, wherein:

the number of persons taking in and taking off the elevator and the operation log information include the number of persons taking in the elevator, the number of persons getting off the elevator, and information indicating an operation rule applicable to operation control of each car belonging to the elevator group when the actual state of each car is obtained and a control parameter set in the operation rule;

the processor executes the second simulation by operating the cars according to the applicable operation rules and the set control parameters.

11. An elevator analysis method, which is an elevator analysis method performed by an elevator analysis system having a processor and a storage device connected to the processor, characterized in that:

the storage device stores: elevator taking in and out number information indicating the actual number of elevator taking in and taking out of each elevator car belonging to the elevator group to be controlled on each floor; running log information indicating an actual state of each car belonging to the elevator group; and the number of passengers, wherein the number of passengers refers to the number of users who appear at the elevator taking position of each floor of the elevator group due to the use of the elevator;

the elevator analysis method comprises the following steps: sequence 1, the processor estimates the number of passengers according to the number information of the passengers on the elevator and the running log information; sequence 2, the processor predicting a future ride size based on the estimated ride size; a sequence 3 in which the processor determines an operation rule applicable to operation control of each of the cars belonging to the elevator group and a control parameter set in each operation rule, based on the predicted number of future passengers; and 4, outputting the determined operation rule and control parameter by the processor.

12. The elevator analysis method according to claim 11, characterized in that:

in the sequence 3 described above, the sequence of the first embodiment,

the processor calculates the probability of destination floors according to the number information of the passengers taking the elevator, wherein the probability of the destination floors is the probability that the destination floors of the users at the elevator taking positions of all floors are the floors;

13. The elevator analysis method according to claim 12, characterized in that:

in the sequence 3 described above, the sequence of the first embodiment,

the processor executes a first simulation a plurality of times while changing the applicable operation rule and the control parameter, wherein the simulation generates users in elevator taking places of each floor and then operates each car in the elevator group according to the number of future passengers and the destination floor probability;

14. The elevator analysis method according to claim 11, characterized in that:

the elevator analysis method further comprises the following sequence:

the processor generates a plurality of users who appear at the elevator taking place of the elevator group due to the use of the elevator, randomly determines the elevator taking place floor of each user, the appearance time of each user and the destination floor of each user, executes a second simulation for operating each car belonging to the elevator group according to the appearance time of each user, the elevator taking place floor and the destination floor, and generates a passenger number estimation model for estimating the number of passengers appearing at the elevator taking place of each floor according to the state of each car, the number of passengers of each car in each floor and the number of passengers going off each car in each floor;

the processor estimates the number of passengers on each floor by applying the actual number of passengers, the actual number of passengers getting off and the actual state of each cage, which are acquired according to the information of the number of passengers getting on and off the elevator and the operation log information, to the number of passengers estimation model;

the processor saving the estimated ride size in the storage device;

in sequence 2, the processor predicts the future number of rides based on the estimated number of rides saved in the storage device.