JPS6219349B2 - - Google Patents

Description

[Detailed Description of the Invention] [Technical Field of the Invention] The present invention provides a highly practical elevator that can automatically measure the traffic volume between floors of the elevator and record this to be useful for operation control and management of the elevator. Regarding the method of recording traffic volume between floors.

[Technical background of the invention and its problems]

Elevator traffic demand often exhibits statistical regularity. The above-mentioned regularity tends to occur at fixed times of the day, such as when starting work, lunch break, and leaving work in an office building, and when excluding special days such as weekends and holidays, the regularity occurs on a daily basis. Similar patterns are repeated. For example, if we examine the number of people moving per unit time using an elevator from the floor where the entrance is located to the floor where the workplace is located, we will see a pattern as schematically shown in FIG. 1, for example, in relation to time. Therefore, by keeping a record of how many people use the elevator at what time and from what floor to what floor, and finding the regularity, it is possible to predict fluctuations in traffic demand for elevators and use the elevators more effectively. It becomes possible to manage the operation. In other words, if we manage and control elevator operations by discovering the above regularity and predicting fluctuations in traffic demand, we can rationally allocate elevator cages to respond to hall calls from multiple floors, increasing efficiency. Good elevator operation and
This makes it possible to transport passengers smoothly.

However, in the past, information such as the weight of passengers in the elevator car (load inside the car), departure floor, car call, hall call, and current time was directly detected, and after this information was registered and stored in a magnetic tape device, etc. A large-scale computer is used to estimate how many people moved from what floor to what floor at what time. Based on this estimation result, the pattern of traffic demand fluctuations is estimated and processed, and this is fed back to the elevator traffic management control.

However, such data collection requires a great deal of time, effort, and expense over several days to several months. Moreover, the amount of data obtained in one day's observation is enormous, and the accumulation and analysis of the data is extremely complicated and difficult. Moreover, it is virtually impossible to conduct the above-mentioned traffic surveys frequently and over long periods of time, and even if the fluctuation pattern changes due to seasonal factors or changes in office hours, for example, it is difficult to It was not possible to detect the problem promptly and reflect it in the elevator operation management control.

[Purpose of the invention]

The present invention was made in consideration of these circumstances, and its purpose is to measure the traffic volume between elevator floors over a long period of time, organize and effectively record this information, and improve the efficiency of the elevator. The object of the present invention is to provide a highly practical method for recording traffic volume between floors of an elevator, which can be effectively used for operation management control.

[Summary of the invention]

The present invention uses the location information of the elevator car, the car call registration information by the passengers who have boarded the elevator car, and the information of at least the number of passengers or the number of passengers getting off the elevator car to identify the floors for each floor pair that the passenger can pass through. Measure the inter-floor traffic volume, classify the measured inter-floor traffic volume into data areas determined corresponding to each floor pair, and integrate the above-mentioned inter-floor traffic volume data according to predetermined time conditions. By doing so, the information is condensed and recorded.

〔Effect of the invention〕

Therefore, according to the present invention, the traffic volume between floors can be easily and accurately measured over a long period of time, and the data can be organized by floor and further categorized and condensed according to temporal conditions and recorded. By compressing the amount of information, it becomes possible to accurately grasp regular patterns of traffic demand fluctuations and effectively utilize them for elevator operation management control.

[Embodiments of the invention]

Hereinafter, one embodiment of the present invention will be described with reference to the drawings.

FIG. 2 is a schematic configuration diagram of an embodiment of an apparatus to which the method of the present invention is applied, in which 1 is a small computer responsible for managing and controlling the operation of elevators, and 2 is a computer that measures the traffic volume between elevator floors using this computer 1. inter-floor traffic measurement device; 3 is an external storage device such as a magnetic tape device;
Reference numeral 4 denotes a clock device that generates time signals such as current time information. The computer 1 generally includes a control device 5 that executes arithmetic processing and various control processes, an internal memory 7 connected to the control device 5 via a bus line 6, and a computer 1 and the measurement device 2, etc. It is composed of a buffer device 8 and the like that transmits and receives data to and from an external device.

The inter-floor traffic measurement device 2 is realized by adopting various measurement methods, but basically, for example, the position information of the elevator car, the elevator car call registration information by the elevator car passenger, and the elevator car From the information on the number of passengers or the number of passengers getting off, which is calculated by weight conversion from the change in internal load, the traffic volume between floors is determined for each floor pair between the departure floor and arrival floor of the elevator cage where passengers can pass through. It is. This inter-floor traffic volume for each floor is determined by, for example, the definite inter-floor traffic volume obtained from the above information and the probabilistic estimated distribution for the remaining undetermined number of people.
The calculation control and processing are performed by the computer 1.

The information obtained by this inter-floor traffic measurement device 2 is, for example, in a data format as shown in FIG. 3, and is input into the computer 1 each time the direction of travel of the elevator car changes. However, it goes without saying that this information transfer of traffic volume between floors may be performed frequently each time there are no passengers in the elevator cage, or it may be performed each time a certain amount of information is accumulated. do not have. The data format shown in FIG. 3 includes departure floor i, arrival floor j, time t when a hall call or basket call occurs,
And the number of people k to be transported from the departure floor i to the arrival floor j
are expressed as a set in binary numbers in time series, and the end data indicated by diagonal lines in the figure is inserted at the end of these series of data to form the data. In addition, when transferring this data, a data request signal (REQ) is output from the counter 1 to the measuring device 2, and after receiving this signal, an ACK signal is issued after the data transfer is completed. It is practically preferable to perform data exchange control and ensure data transfer.

Now, the internal memory 7 of the computer 1 is composed of, for example, a random access memory (RAM), and the memory 7 has a data structure area prepared for each day of the week as shown in FIG. ing. The inter-floor traffic data search and update explained below uses the data structure shown in Figure 4 above.
This is done by performing operations for each day of the week indicated by . Therefore, for example, a data area having a one-dimensional array structure of x ^m , y ^m , and z ^m given as elements of 1 word and 16 bits starts at addresses x ^m _p , y ^m _p , and z ^m _p , respectively. These have mutual address correspondence. The data area x ^m stores information regarding the floor pair ij determined by the departure floor i and the arrival floor j, and information regarding ^the time t ^m . The above-mentioned rank pair ij is expressed by a ring of pointers that starts from the element x ^m _(A) and returns to the above element x ^m _(A) at the end, and the above time t
information is now being provided. Furthermore, the data area y ^m stores predicted value data of the average number of people moving during the time period corresponding to the pointer element x ^m _(i) . The data stored in this data area y ^m provides information regarding the traffic demand pattern of the elevator system. The data area z ^m stores new empirical information, which will be described later, for updating the knowledge (information) represented by x ^m and ^ym .

By the way, the floor pair (floor pair) indicated by the above pointer ring is expressed as follows. That is, this is done by assigning a series of numbers to all the floors serviced by the elevator. For example, if there are service floors from the 1st floor to the 5th floor, and the inter-floor traffic volume is to be measured for all of them, for example, as shown in Figure 5a, the floor pairs j, i
Assign numbers h (h=1, 2, 3 to 20) in order. Note that this number h and the floor pair i, j are one to one.
The order of numbering is not particularly limited as long as it corresponds to The floor pair may be removed from the pointer ring and numbered as shown in FIG. 5b. Furthermore, there are no particular limitations on how to value m. In the following explanation, since the explanation becomes unnecessarily complicated as the number of floors increases, the number of service floors is assumed to be "3" and the pointers are numbered as shown in Figure 5c. shall be taken as a thing.

The data area x ^m shown in FIG. 4 is divided and set corresponding to the pointer h determined in this manner. Specifically, when an elevator serving the third floor collects data in four time periods as shown in FIG. 6, h=1,
2, 3, 4, 5, 6, and one data amount A is 24
The data structure of x ^m in words is shown in FIG. One area delimited by the pointer corresponds to one floor pair. In this example, x ^m ₍₅₎ , x ^m ₍₆₎ , x ^m ₍₇₎ correspond to floor pair h=2.

Therefore, in order to determine the area corresponding to a certain floor pair h, the pointer is sequentially transmitted from the pointer number position x ^m _(A) and the number of times the pointer is transmitted is counted. This is done by For example, the area corresponding to the floor pair h = 2 is defined as the area corresponding to the next pointer position l = 8 indicated by the pointer indicated at l = 4 initially transmitted from x ^m _(A). It will be done. As a result, the above l
From the next word after =5, the second pointer position l=8
Addresses x ^m ₍₅₎ to x ^x ₍₈₎ of x, which correspond to the above, are shown as an area with the number of floors h=2. In general, the area corresponding to h in a specified m starts from x ^m _(A) and starts from the next word after passing the pointer (h-1) times, starting from h
It will be shown as the word that was transmitted twice.

Further, data t ^m ₁ shown in FIG. 7 indicates the time [0, t ^m ₁ ] from time 0 to time t ^m ₁ , and information regarding that time is given as data y ^m _h1 and z ^m _h1 . Further, data t ^m ₂ indicates time [t ^m ₁ , t ^m ₂ ], data t ^m ₃ indicates time [t ^m ₂ , t ^m ₃ ], and information regarding these times is data y ^m _h2 , z ^m _h2 , y ^m _z
3 ,z
^m _h3 . Furthermore, data y ^m _h4 , z ^m
_h4
is information regarding the time (t ^m ₃ , T ^m ) from time t ₃ to time 0 (=T ^m ) of the next day.

Note that this time T ^m does not necessarily have to be set to midnight. Further, the number of pointers in the data structure shown in FIG. 7 is as schematically shown in FIG. 8.

As mentioned above, regarding x ^m , its connection relationship is determined by the pointer, and thereby,
The data areas of y ^m and z ^m are respectively specified. As mentioned above, the area y ^m specified by the above pointer stores information on the amount of traffic (number of people) expected to move at that time with respect to the floor pair h, and the area y m specified by the pointer The area ^m stores the amount of traffic actually moving using the elevator under the same conditions as above. As a result, the inter-floor traffic measured by the inter-floor traffic measuring device 2 and classified by floor pairs with passability is classified according to predetermined temporal conditions, for example, days of the week, and corresponds to each floor pair. The information is classified by predetermined time period and recorded, for example, in a sequentially integrated manner, in a corresponding area of the data area determined by the above-mentioned pointer and managed by the above-mentioned pointer.

The value of m is generated by the clock device 4, and is used to indicate the day of the week. And for example 16
In hexadecimal representation, (00) is Sunday and (01) is Monday
(06) is expressed as Saturday, etc.

Next, the processing action of the method of the present invention will be explained.

FIG. 9 shows the control flow of the elevator by the computer 1, and the information recording processing routine is incorporated into this main control routine.
That is, when the control program is started, first the elevator control procedure is initialized in step 11, and then the data structure shown in FIG. 4 is initialized in the next step 12. after that,
After the value of m is determined (designated) in step 14 via repeat point 13, steps 15 and 16
A processing routine for performing the elevator control procedure shown in is repeatedly executed. Note that the value of m indicates information about the day of the week. Now, the control routine of this method incorporated into such a processing routine first performs step 17 based on the determined value of m.
Reference is made to the input buffer at , and determination section 18 determines the first word of the input buffer. This determination is to check whether new inter-floor traffic measurement data has been provided by the inter-floor traffic measurement device 2. When new data is available, the process jumps to subroutine 19 and registers the data in the data array z ^m shown in FIG. 4. The presence or absence of this new data is checked by referring to the information in the input buffer 8 connected to the inter-floor traffic measurement device 2.

Thereafter, the determination unit 20 determines whether there is a prediction request to predict how much traffic demand there will be for a certain pair of floors at a certain time in the elevator control procedure. In response to this prediction request, necessary information is retrieved from the data structure shown in FIG. 4 in the subroutine 21, and a prediction calculation to be described later is executed. After this prediction calculation is performed, or if there is no prediction request, the control procedure shown in step 16 is performed, and then the determination unit 22 determines whether or not it is okay to update the data. If the data can be updated based on this determination, the data structure update procedure shown in FIG. 4 is performed in subroutine 23. This data update is performed by changing the contents of arrays x and y based on new data registered in array z in the data structure shown in FIG.

The above is a general data processing procedure in this method.

Now, the initialization of the data structure shown in step 12 is performed, for example, as shown in FIG. This initialization is performed by setting all the contents of the arrays y and z to "0", and first, step 2
In step 4, the value of m is set to "0", and then in step 25, the value of l is set to "1", and in steps 26 and 27, y ^m (specified by the value of l) is set _. The values of z _{m (l)} and z ^m _(l) are each set to "0". Then, the determination unit 28 determines whether the value of l has reached the value of the magnitude A, and step 2
At step 9, the value of l is incremented and this is repeated. Thereafter, the determining section 30 checks the value of m, and if the value does not reach "7", the value of m is incremented at step 31, and the above process is repeated. As a result, the fourth
All contents of the data structure shown in the figure are set to "0", and its initialization is completed.

In this state, if the first word of the data stored in the input buffer 8 is read in step 12, if there is no new data in the input buffer 8, the end signal of the previous data, e.g. END Since OF FILE remains, it is determined whether new data exists. however,
This end signal is given after the last data when the inter-floor traffic measuring device 2 outputs new data and sequentially stores them in the input buffer 8.

Now, the data registration shown in subroutine 19 is performed according to the routine shown in FIG.
That is, first, after initializing the value of n to "1", in step 32, the first four
Read the words i _n , j _n , t _n , k _{n ,} respectively. Next, in step 33, the number h corresponding to the floor pair i _n , j _n is searched, and then in step 34, the array number h corresponding to the floor pair h and the time t ^m is found, indicating the area of the array z ^m . Search for box. Since the process of finding the array number box of z ^m is somewhat complicated, this will be explained later with reference to FIG. 12. Thereafter, in step 35, the number of people moving k is added to the area z ^m (box) specified by the box, and then the contents of the input buffer 8 are shifted to the left by four words. Through the processing of step 36, the input buffer 8 is prepared to receive the next new data. Thereafter, in step 37, the value of n is incremented and the above process is repeated.
Thereafter, when all the data for that time has been processed, an end-of-file signal remains in the input buffer 8.

Now, in the processing routine shown in FIG.
The box heading is done as follows. First, in step 38, it is determined whether the floor pair number h corresponds to the elevator to be controlled. If it is not applicable, step 3
After steps 9 and 40, the values p=0 and q=A are set, and the process moves to the next step. Also, the above number h
If the object is a target, the routine shown in steps 41 to 47 follows the pointer shown in FIG. 4. That is, in steps 41, 42, and 43, p is set to "0", q is set to "x ^m _(A) ", and r is set to "1", and a variable for searching the data area corresponding to h is defined. be done. However, the above q is the current position of the pointer, p is the previous value, and r is the number of times the above q has been changed starting from the initial position x ^m _(A) .
In other words, each value is defined as a value indicating the number of times the pointer has been crossed. The determination unit 44 then determines the value of r defined above and executes the pointer tracking. In other words, the process continues in step 45 until the number of pointer traverses reaches its maximum value h.
-47 routines are driven. Steps 45-4
7, the value of q is set to the content of p, and x ^m _(q) is set to the content of q. Thereafter, the value of r is incremented, and by repeatedly executing this process, the pointer position is sequentially traced by the pointer. By repeating this processing routine h times, the data area corresponding to the floor pair h is determined. after that,
By executing step 48, "1" is added to the value of p, and the starting point p and ending point q of the area are determined. For example, in the example shown in FIG. 8, if h=2 is given, the values p=5 and q=8 will be obtained, respectively.

Thereafter, in the next step 49, it is determined whether the value of p is equal to q. Here, p=
When the determination q is made, the data area corresponding to the floor pair h is only one word, so in step 50, the value of box is determined as p=q. Also,
If the determination result of p=q is not obtained, the next determination unit 51 determines whether the value of p is smaller than q. If the conditions are not met, this will be considered an abnormal situation.
At step 52, appropriate appeal processing such as setting an abnormality flag is performed.

Now, when the above condition p<q is satisfied,
This time, the following conditions are sequentially checked in the determination units 53, 54, and 55. That is, first, the determination unit 5
In step 3, the conditional determination that 0≦t≦x ^m _(p) is made,
When this condition is met, in step 56,
The value of box is determined to be p. Further, if the above conditions are not satisfied, the determination unit 54 determines that x
A conditional judgment is made that ^m _(q-1) t<T ^m , and when this condition is satisfied, the value of box is determined to be q in step 57. However, if the above conditions are not satisfied, then the determination unit 55
In this step, the condition x _(p) t<x ^m _(q-1) is determined, and if this condition is also not satisfied, this is determined to be an abnormality, and the abnormality flag processing in step 52 is performed.

Next, in the judgment section 55, if the condition is satisfied, the value of box is initialized to (p+1) in step 58, and the judgment sections 59, 60
, p<box<q and x ^m _(bpx-1) ≦t<x
^m _(bpx)
At step 61, the value of the box is incremented, and this is repeated to determine the value of the box. Through such a processing routine, the value of the array array number box having information on the floor pair h at time t is found and determined. Then, the processes of steps 35 and 36 shown in FIG. 11 are performed subsequently.

Next, the determination of the prediction request in step 20 shown in FIG. 9 is performed as follows. This determination is made by providing a data area in the memory of the computer 1 to store, for example, the value of h and the value of t, and
As shown by two decision steps 62 and 63 in FIG. 3, it is determined whether h=0 and at the same time whether t ^m =0. Then, when the above two conditions are simultaneously satisfied, that is, when the value of h and the value of t are both 0, it is determined that there is no prediction request. Further, if the value of either h or t is not 0, it is determined that there is a prediction request.

In the routine shown in step 21 in which processing is executed in response to such a prediction request, the predicted value is calculated as shown in FIG. That is, in this routine, first, in step 64, p,
The values of box and q are respectively taken in. Thereafter, the determination units 65 to 68 compare the values of p, box, and q. This comparison is performed by comparing whether the values of p, box, and q are the same, or their magnitude relationship. If this comparison condition is not satisfied, abnormality flag processing is performed in step 69.

Specifically, first, in step 65, p=
A determination that box=q is made. Then, when this condition is satisfied, in step 70,
The predicted value Wev is determined as Wev=y ^m _(bpx) /T ^m . This predicted value Wev is
It has the meaning of the predicted value of the average number of people moving per unit time at time t. Further, if the condition determined by the determination unit 65 is not satisfied, then the determination unit 66
A conditional determination is made that p=box<q. If this condition is satisfied, the predicted value described above is obtained in step 71 by the calculation process Wev=y ^m _(bpx) /x ^m _(bpx) . Further, if each of these conditions is not satisfied, then the determination unit 67 determines that p<box=
A conditional determination of q is performed. If this condition is met, in step 72, Wev=y ^m _(bpx) /{x ^m _(bpx) - x ^m _(b
px-1) } The predicted value is calculated by the operation. Further, if this condition is not satisfied, the determining unit 68 makes a conditional determination that p<box=q, and when this condition is satisfied, in step 73, Wev=y ^m _(bpx) /{T ^m -x ^m _(bpx-1
) } to find the predicted value. However,
If all of the above-mentioned conditions are not met, these values cannot originally occur in the control routine of this system, so this is determined to be an abnormality.

In addition, x _(bpx) used in these predicted value calculations
The value indicates time information, for example, with 8:00 a.m. as the reference time "0" and the time up to 8:00 p.m. the next day expressed in seconds. In this example, the relative time difference from 8 a.m. is 9
They will be given at 1:30 p.m., 1:00 p.m., and 6:00 p.m., respectively. Therefore, the arithmetic expressions shown in steps 70 to 73 above are generally given as the time period (seconds) for the value of m when corresponding to Wev=y ^m _(bpx) /box.

Next, the data update determination (step 22) and data update routine (step 23) in the routine shown in FIG. 9 will be explained. A determination as to whether data update is possible is made by considering whether or not the conditions described above are satisfied. For purposes of explanation, it is assumed here that the traffic demand for elevators becomes extremely low after a certain time, such as at night, and that this state continues for several hours. Under such assumed conditions, it is possible to determine whether the above data update is possible by referring to the current time and determining whether that time has passed a predetermined time to. becomes possible.

Therefore, in this case, for example, as shown in FIG. 15, the determination unit 74 first determines whether the current time is within the range from time t ₀ to time (t ₀ +100). . The time width "100" shown at the above time (t ₀ + 100) is set in order to execute the update routine shown in step 23 only once during that 100 seconds and avoid unnecessary repetition. It is. If the above conditions are not met, the data flag is set to "0" in step 75, and the process proceeds to the next step without updating the data. If the above conditions are met, the determination unit 76 refers to the value of the data flag and leads the process to the data update step only when the data flag is "0". In this case, step 77
The flag is set to "1" at the time t, and the data is controlled so that it is updated only once at a certain time t.

In the data update step 23 activated by such a determination, for example, as shown in FIG. 16, the update process is executed by exponential smoothing of the data. That is, first, in step 78, the value of l is initialized to "1". Thereafter, the determination section 79 determines the value of l, and the data update processing in steps 80 and 81 is repeatedly executed. In step 80, the weighting coefficient of the new data relative to the old data is set to a, and data update processing is performed as follows: y _(l) = a x z _(l) + (1-a) x y _(l) . The above coefficient a is
It is given arbitrarily in the range 0≦a≦1, and a
=0 is a special condition when new data is ignored, and a=1 is a special condition when old data is ignored. After this process is executed corresponding to the given l, the value of l is incremented in step 81, and the value of y is The update process will be performed.

As described above regarding the data recording processing function according to the present invention, according to the present invention, the traffic demand of elevators (traffic volume between floors) can be automatically and over a long period of time without using special sensors. The data is measured by floor, classified by day of the week and time of day, which are temporal conditions, and sequentially added to the corresponding data area, so the amount of information can be effectively reduced. information can be recorded. Moreover, since the inter-floor traffic volume determined for each floor pair in a predetermined time period is averaged to obtain characteristic data and recorded, it is possible to perform even more effective information reduction. Therefore, the capacity of the memory for storing this information can be significantly reduced, and it becomes possible to store it using, for example, several kilowords of RAM. By the way, if the measured data is stored as is, as has been done in the past, the amount of information for one day alone would amount to about 20 kilowords, and this would be accumulated sequentially over several months. A huge amount of memory capacity is required to store information. Moreover, the information organization and processing becomes extremely complicated, so it is extremely impractical. Therefore, it can be said that the recording method according to the present invention is extremely useful. Furthermore, since the data is compressed and recorded as described above, it has great practical effects, such as being able to be used simply and effectively in elevator operation management control.

Note that the present invention is not limited to the above embodiments. For example, the day of the week information m may be divided into two types, weekdays and holidays.In this case, Saturdays may be set as either weekdays or holidays depending on the nature of the building where the elevator is installed. . When m is determined in this way, m
It is sufficient to simply set the possible values of to "0" and "1", and it is possible to significantly reduce calculation processing and storage area for the data. Also, depending on the specifications, the time zone settings may be set differently depending on the value of m (weekdays and holidays) as shown in Figure 17. In this case, the size of A may be changed to m. It is desirable to set it variably depending on the value, and in the explanation of the operation of the above-mentioned embodiment, A may be read as A ^m and the processing routine thereof may be executed. Then, the data structure in memory is
As shown in FIG. 8, the arrays of x ^m _(l) , y ^m , and z ^m may be prepared differently depending on the value of m. Incidentally, FIG. 19 shows an example of the value of A ^m that is read by the value of m, and for example, the conversion table may be prepared as a table in the memory. In other words, in the present invention, since the pointer specifies the position of the information to be viewed next,
As described above, it is possible to easily set an arbitrary time zone according to time conditions such as the day of the week. Further, in the above-described embodiment, the data obtained by the inter-floor traffic measurement device 2 is periodically retrieved and input, but this may be performed by forced interruption processing. Alternatively, information processing may be performed using a dedicated computer,
This may be performed using a time change machine for group management control for a plurality of elevators. Furthermore, of course, the type of elevator, its operating speed,
There are no particular limitations on the number of service floors, etc.
In summary, the present invention can be implemented with various modifications without departing from the gist thereof.

[Brief explanation of the drawing]

Fig. 1 is a diagram showing an example of a variation pattern of inter-floor traffic volume due to elevators, Fig. 2 is a schematic configuration diagram of an embodiment of an apparatus to which the method of the present invention is applied, and Figs. 3 to 8 are diagrams according to the present invention. 9 to 16 are diagrams showing an example of the information processing control flow according to the present invention, and FIGS. 17 to 19 are diagrams each showing a modification of the present invention. be. 1... Calculator, 2... Inter-floor traffic measurement device, 1
1 to 81...Processing steps.

Claims (1)

[Scope of Claims] 1. The position information of the elevator car, the information of the car call by the passenger of the elevator car, and at least the number of people getting on or getting off at the stop floor of the elevator car are measured, and based on these information, the number of people getting on or getting off the car is determined. When calculating and recording the inter-floor traffic volume in a predetermined data area, the car call by the passenger of the elevator car is determined based on the information on the stop floor of the elevator car where the passenger boarded and alighted, and the position of the elevator car. From this information, find floor pairs consisting of a departure floor and an arrival floor that the passenger can pass through, and calculate the number of boarding or alighting at the stop floor of the elevator car for these floor pairs that the passenger can pass through. The inter-floor traffic volume of the passengers is classified into each possible floor pair by allocating the number of people, and the inter-floor traffic volume classified by these floor pairs is determined according to predetermined time conditions. The inter-floor traffic volume classified by floor pair is consolidated for each time period, and the inter-floor traffic volumes classified by floor pair are summarized, and each floor pair with possible traffic is condensed and rearranged according to the predetermined temporal conditions. The data area for recording the traffic volume between floors is divided into a plurality of partition areas corresponding to each floor pair, and the partition position of the next partition area is set at the position where the partition position of each partition area in the data area is determined. Each section is managed by being associated with each floor pair based on the connection of these pointers, and each section is classified according to the floor pair, and according to a predetermined time condition. A method for recording inter-floor traffic volume of an elevator, characterized in that the inter-floor traffic volume is recorded respectively, reduced and organized according to the following. 2. The predetermined time condition is given by the day of the week when the elevator is operated, by weekdays and holidays, and by the time of the day. How to record quantities.