JP6161386B2 - Queue estimation system, average waiting time calculation method, queue length calculation method, average waiting time calculation program, and queue length calculation program - Google Patents

Description

  The present invention relates to a queue estimation system, an average waiting time calculation method, a queue length calculation method, an average waiting time calculation program, and a queue length calculation program.

  Japanese Unexamined Patent Application Publication No. 2007-128386 (Patent Document 1), which is the background art of the present invention, “calculates the user's waiting time by grasping the time difference from the store entry time to the use start time, "Provides waiting time for ATM stores".

  Japanese Patent Laid-Open No. 07-044424 (Patent Document 2) states that “the number of arrivals and departures of a call is counted (step 401), and the number of arrivals before x hours and the current number of departures are calculated each time a departure occurs. Comparison is made (step 402), and the case where the number of arrivals before x hours is larger is counted (steps 403 and 404), and after the measurement period is over (step 405), the number of calls that have been withdrawn within the measurement time The number of calls counted in (1) is divided to obtain an estimated value of the probability distribution of the waiting time (step 406) ”.

  Japanese Patent Application Laid-Open No. 09-035125 (Patent Document 3) describes that “the waiting time of the customer is calculated from the difference between the time of visiting the store and the time of accepting the use ticket”.

JP 2007-128386 A Japanese Patent Application Laid-Open No. 07-044424 Japanese Patent Laid-Open No. 09-035125

  In the conventional method, in order to calculate the queue length (number of people) and waiting time, in addition to the service log provided by the external terminal, the customer must acquire the time at the end of the queue separately. did not become.

  The object of the present invention is to obtain an average waiting time and waiting time required to use an external terminal based on a service log provided by the external terminal without separately acquiring the time at which the customer is queued at the end of the queue. It is to provide a technique that makes it possible to provide the length of a matrix for each time.

  Of the inventions disclosed in the present application, the outline of typical ones will be briefly described as follows.

  One embodiment of the present invention includes an average service time calculation unit that calculates the time from the service start time to the service end time included in the provided service log as an average service time. Further, an arrival time calculation for calculating an average arrival time by dividing the service time calculated based on the service start time and the service end time in the log by the total number of the services processed during the service time. Part. Furthermore, a queue estimation unit that calculates an average waiting time based on the calculated average service time and the average arrival time.

  In another embodiment, an average service time calculation unit that calculates a time from a service start time to a service end time included in the provided service log as an average service time is provided. Further, an arrival time calculation for calculating an average arrival time by dividing the service time calculated based on the service start time and the service end time in the log by the total number of the services processed during the service time. Part. In addition, a queue estimation unit that calculates a queue length based on the calculated average service time and the average arrival time.

  In another embodiment, the average service time calculation unit includes an average service time calculation step of calculating the time from the service start time to the service end time included in the provided service log as an average service time. The arrival time calculation unit divides the service time calculated based on the service start time and the service end time in the log by the total number of the services processed during the service time, thereby obtaining an average arrival time. An arrival time calculating step for calculating. The queue estimation unit further includes an average waiting time calculating step of calculating an average waiting time based on the calculated average service time and the average arrival time.

  In another embodiment, the average service time calculation unit includes an average service time calculation step of calculating the time from the service start time to the service end time included in the provided service log as an average service time. The arrival time calculation unit divides the service time calculated based on the service start time and the service end time in the log by the total number of the services processed during the service time, thereby obtaining an average arrival time. An arrival time calculating step for calculating. The queue estimation unit further includes a queue length calculation step of calculating a queue length based on the calculated average service time and the average arrival time.

  In another embodiment, an average service time calculation step of calculating, as an average service time, a time from a service start time to a service end time included in the provided service log by the average service time calculation unit is queued. Let the estimation system computer run it. The arrival time calculation unit divides the service time calculated based on the service start time and the service end time in the log by the total number of the services processed during the service time, thereby obtaining an average arrival time. The computer of the queue estimation system is caused to execute an arrival time calculating step for calculating. Further, the queue estimation unit causes the computer of the queue estimation system to execute an average waiting time calculating step of calculating an average waiting time based on the calculated average service time and the average arrival time.

  In another embodiment, an average service time calculation step of calculating, as an average service time, a time from a service start time to a service end time included in the provided service log by the average service time calculation unit is queued. Let the estimation system computer run it. The arrival time calculation unit divides the service time calculated based on the service start time and the service end time in the log by the total number of the services processed during the service time, thereby obtaining an average arrival time. The computer of the queue estimation system is caused to execute an arrival time calculating step for calculating. Further, the queue estimation unit causes the computer of the queue estimation system to execute a queue length calculation step of calculating a queue length based on the calculated average service time and the average arrival time.

  Among the inventions disclosed in the present application, effects obtained by typical ones will be briefly described as follows.

  According to an embodiment of the present invention, it is necessary to use an external terminal based on a service log provided by the external terminal without separately acquiring the time at which the customer is queued at the end of the queue. The average waiting time and queue length can be provided for each time.

It is a figure which shows the outline | summary of the structural example of the queue estimation system in one embodiment of this invention. It is a figure which shows the outline | summary of the structural example of the log in one embodiment of this invention. It is a figure which shows the outline | summary of the structural example of the average service time table in one embodiment of this invention. It is a figure which shows the outline | summary of the structural example of the output information storage table in one embodiment of this invention. It is a figure which shows the example of the estimation result display screen in one embodiment of this invention. It is a figure which shows the outline | summary of the queue estimation process in one embodiment of this invention. It is a figure for demonstrating the average service time calculation process in one embodiment of this invention. It is a figure for demonstrating the complementary process in the average service time calculation process in one embodiment of this invention. It is a figure which shows the outline | summary of the average arrival time calculation process in one embodiment of this invention. It is a figure for demonstrating the process which calculates the average arrival time of the service in one embodiment of this invention. It is a figure for demonstrating the complementation process in the average arrival time calculation process in one embodiment of this invention.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Note that components having the same function are denoted by the same reference symbols throughout the drawings for describing the embodiment, and the repetitive description thereof will be omitted.

<System configuration>
FIG. 1 is a diagram showing an outline of a configuration example of a queue estimation system according to an embodiment of the present invention.

  In FIG. 1, a queue estimation system includes a log acquisition unit 101, an average service time calculation unit 102, an average arrival time calculation unit 103, a storage unit 105, a queue estimation unit 106, an output unit 107, and an input. Part 108.

  The average arrival time calculation unit 103 includes a matrix break detection unit 109 and an arrival time calculation unit 110.

  The queue estimation system is implemented with predetermined hardware and software. For example, the queue estimation system includes a processor, a memory, and the like, and causes the computer to function as the queue estimation system by executing a program on the memory by the processor.

  The log acquisition unit 101 acquires a log (described later, FIG. 2) from an external terminal (not shown) (for example, ATM (Automated Teller Machine)) via the network 104.

  The average service time calculation unit 102 stores the average service time in an average service time table (described later, FIG. 3) based on the acquired log. The average service time calculation unit 102 calculates an average service time for each representative time based on the acquired log, and stores the calculated average service time in association with the representative time in an output information storage table (described later, FIG. 4). To do.

  The average arrival time calculation unit 103 performs an average arrival time calculation process (described later in FIG. 9), and stores the calculated average arrival time for each representative time in the output information storage table.

  The queue estimation unit 106 calculates the average queue length and the average waiting time from the output information storage table based on the average service time and the average arrival time, and calculates the average queue length and the average waiting time. A certain average waiting time is stored in the output information storage table.

  The storage unit 105 stores a log, an average service time table, and an output information storage table.

  The output unit 107 displays an estimation result display screen (described later, FIG. 5) based on the information acquired from the output information storage table.

  The input unit 108 corresponds to an input device such as a mouse that receives an input to the queue estimation system.

<Log configuration>
FIG. 2 is a diagram illustrating an outline of a configuration example of the log 201 according to the embodiment of this invention. The log 201 includes data that records the contents of a service for an operation received by an external device, the date and time when the external terminal provided the service, and the like.

  The log 201 has data items such as [log ID], [date], [time], [service ID], [processing content], [terminal ID], and [remarks].

  [Log ID] indicates a symbol for identifying the log 201. [Date] indicates the date when the external terminal provided the service. [Time] indicates the time when the external terminal provided the service. [Service ID] is a symbol assigned to each customer, and the same symbol is assigned to the same customer. [Processing content] indicates the content of the service provided by the external terminal (for example, cash payout or balance inquiry in ATM). [Terminal ID] indicates a symbol for identifying the external terminal that has performed the process. [Remarks] stores information for distinguishing whether the corresponding [Time] is a service start time or a service end time.

<Average service time table>
FIG. 3 is a diagram showing an outline of a configuration example of the average service time table 301 according to the embodiment of the present invention.

  The average service time table 301 includes data items such as [service ID], [representative date], [representative time], [average service time], and [terminal ID].

  [Representative date] indicates the date on which the representative time is calculated. [Representative time] indicates a time at which the external terminal is estimated to have provided a service. [Average service time] indicates the average service time for each customer.

<Output information storage table>
FIG. 4 is a diagram showing an outline of a configuration example of the output information storage table 401 in the embodiment of the present invention.

  The output information storage table 401 has data items such as [date], [representative time], [average service time], [average arrival time], [queue length], and [average waiting time].

  [Date] indicates the date when the external terminal provided the service. [Representative time] indicates a time at which an external terminal is estimated to provide a service.

  [Average arrival time] indicates the average time of the visit of the customer using the external device.

  [Queue Length] indicates the average number of customers on the external device.

  [Average waiting time] indicates an average time that the customer must wait to use the external device.

<Example of estimation result display screen>
FIG. 5 is a diagram showing an example of an estimation result display screen according to the embodiment of the present invention.

  In FIG. 5, the output unit 107 (for example, a display) displays a graph with the horizontal axis representing the representative time and the vertical axis representing the average waiting time and queue for each branch where the external device is installed. In the example of FIG. 5, the average waiting time and the average queue of the external devices installed in the branch A are displayed in time series. In FIG. 5, the solid line indicates the average waiting time for each representative time. The broken line indicates the queue length for each representative time.

  When the branch selection button 901 accepts the selection, the output unit 107 switches the display to a graph representing the state of another branch.

<Queue estimation process>
FIG. 6 is a diagram showing an outline of the queue estimation process in one embodiment of the present invention.

  First, in S <b> 201, the log acquisition unit 101 acquires the log 201 via the network 104 and stores it in the storage unit 105.

  Next, in S202, the average service time calculation unit 102 performs an average service time calculation process (described later, FIG. 7 and FIG. 8) and stores the calculated average service time for each representative time in the output information storage table 401. .

  Next, in S <b> 203, the average arrival time calculation unit 103 performs an average arrival time calculation process (described later in FIG. 9), and stores the calculated average arrival time for each representative time in the output information storage table 401.

Next, in S204, the queue estimation unit 106, based on the average service time (Ts) and average arrival time (Ta) acquired from the output information storage table 401, the queue length (Lw) and the average wait time. The queue is calculated by calculating the time (Tw) based on the following formula, and the calculated queue length (Lw) and average waiting time (Tw) are stored in the output information storage table 401.
Lw = ρ / (1-ρ) = Ts / (Ta−Ts)
ρ: Service utilization rate (ρ = Ts / Ta)
Tw = Lw × Ts = Ts × Ts / (Ta−Ts)
Next, in S205, the result is output. More specifically, the output unit 107 acquires the date, representative time, queue length, and average waiting time from the output information storage table 401, and displays an estimation result display screen (described above, FIG. 5) based on the acquired information. Display and finish the queue estimation process.

<Average service time calculation process>
7 and 8 are diagrams for explaining the contents of the average service time calculation processing in the embodiment of the present invention.

  As shown in FIG. 7, each service (services n to n + 2 in the figure) is continuously performed by the external device from the service start time to the service end time.

  In the average service time calculation process, the average service time calculation unit 102 acquires the log 201 from the storage unit 105. Then, the average service time calculation unit 102 calculates the elapsed time from the service start time to the service end time as the average service time for each service ID. Further, the average service time calculation unit 102 calculates an intermediate time from the service start time to the service end time as a representative time for each service ID. The average service time calculation unit 102 stores the calculated average service time, the representative time, the service ID, and the terminal ID in the average service time table 301 in association with each other.

  Here, the representative times calculated by the average service time calculation unit 102 do not exist at equal intervals. Therefore, the average service time calculation unit 102 performs a process of complementing the representative time for which there is no corresponding average service time using an interpolation process or a spline curve every minute. As a result of this complementing process, as shown in FIG. 8, data with the horizontal axis representing the representative time and the vertical axis representing the average service time can be obtained. Then, the average service time calculation unit 102 stores the representative time after the complementing process and the average service time in association with each other in the output information storage table 401.

<Average arrival time calculation process>
FIG. 9 is a diagram showing an outline of the average arrival time calculation processing in the embodiment of the present invention.

  First, in step S <b> 701, the matrix break detection unit 109 of the average arrival time calculation unit 103 acquires the log 201 and calculates a free time between each service of the acquired log 201. More specifically, the matrix break detection unit 109 acquires the log 201 from the storage unit 105, and calculates each idle time based on the time from the service end time of each service to the service start time of the next service.

  Next, in S702, the matrix break detection unit 109 detects breaks. More specifically, the matrix break detection unit 109 determines that the service is interrupted in this free time when each free time calculated in S701 is equal to or longer than a predetermined time.

  Next, in S703, the arrival time calculation unit 110 determines the service start times of all the service IDs of services that exist immediately after it is determined that the service is interrupted (that is, all estimated to have no waiting time). Service start time).

Next, in S704, the arrival time calculation unit 110 calculates the average arrival time (Ta) of the service. Step S704 will be described with reference to FIG. This average arrival time (Ta) is the service start time 1102 (Tk) of the service existing immediately after it is determined that the service is next interrupted from the service start time 1101 (Tk (a)) of the service acquired in S703. It is calculated by dividing the time until (b)) by the total number N (a, b) of services processed during that time. In addition, the arrival time calculation unit 110 performs an intermediate operation from the service start time 1101 (Tk (a)) to the service start time 1102 (Tk (b)) of the service that exists immediately after it is determined that the service is next interrupted. Is calculated as the representative time 1103.
Ta = (Tk (b) −Tk (a)) / N (a, b)
Here, the representative times calculated by the arrival time calculation unit 110 do not exist at equal intervals. Therefore, in S705, the arrival time calculation unit 110 performs a process of complementing the representative time for which there is no corresponding average arrival time every minute by using an interpolation process, a spline curve, or the like. As a result of this complementing process, as shown in FIG. 11, data with the horizontal axis representing the representative time and the vertical axis representing the average arrival time can be obtained. Then, the arrival time calculation unit 110 associates the representative time after the complementing process with the average arrival time and stores them in the output information storage table 401.

  Here, S202 corresponds to an average service time calculation step, S203 corresponds to an arrival time calculation step, and S204 corresponds to an average wait time calculation step and a queue length calculation step.

<Effects of the present embodiment>
According to the present embodiment described above, the average service time and average arrival time are calculated based on the service start time and service end time included in the log 201, so that customers are arranged at the tail of the queue. Without acquiring the time separately, the average waiting time and queue length required for using the external terminal can be provided for each time.

  Further, the time from the service start time of the service existing immediately after the idle time determined to be interrupted to the service start time of the service existing immediately after the idle time determined to be interrupted next is serviced It can be calculated as time.

  Further, by storing the representative time in association with the service time at the representative time in the output information storage table 401, a graph with the horizontal axis representing the representative time and the vertical axis representing the average waiting time can be displayed.

  As mentioned above, although the invention made | formed by this invention was concretely demonstrated based on embodiment, this invention is not limited to the said embodiment, It can change variously in the range which does not deviate from the summary. Needless to say.

  The present invention does not separately acquire the time at which the customer is queued at the end of the queue, and based on the service log provided by the external terminal, the average waiting time and the queue required for using the external terminal It has the excellent effect of providing length for each time, and is also applicable to queue estimation system, average waiting time calculation method, queue length calculation method, average waiting time calculation program, and queue length calculation program can do.

101 log acquisition unit 102 average service time calculation unit 103 average arrival time calculation unit 104 network 105 storage unit 106 queue estimation unit 107 output unit 108 input unit 109 matrix break detection unit 110 arrival time calculation unit 201 log 301 average service time table 401 Output information storage table 901 Branch selection button

Claims (9)

An average service time calculation unit that calculates the time from the service start time to the service end time included in the provided service log as an average service time;
Arrival time calculation for calculating the average arrival time by dividing the service time calculated based on the service start time and the service end time included in the log by the total number of the services processed during the service time And
A queue estimation system comprising: a queue estimation unit that calculates an average waiting time based on the calculated average service time and the average arrival time. The queue estimation system according to claim 1,
If the idle time from the service end time of the service to the service start time of the next service is equal to or longer than a predetermined time, the matrix further includes a matrix break detection unit that determines that the service is interrupted during the idle time. And
The arrival time calculation unit is configured to determine whether the queue is determined to be interrupted next from the service start time of the service existing immediately after the idle time when the matrix break detection unit is determined to be interrupted. A queue estimation system for calculating a time until the service start time of the service existing immediately after as the service time. The queue estimation system according to claim 2,
The average service time calculation unit associates a representative time that is an intermediate time from the service start time to the service end time and the service time at the representative time, and stores them in an output information storage table. . The queue estimation system according to claim 3, wherein
Based on the representative time acquired from the queue estimation system and the average waiting time, the queue estimation further includes an output unit that displays a graph with the horizontal axis representing the representative time and the vertical axis representing the average waiting time. system. An average service time calculation unit that calculates the time from the service start time to the service end time included in the provided service log as an average service time;
Arrival time calculation for calculating the average arrival time by dividing the service time calculated based on the service start time and the service end time included in the log by the total number of the services processed during the service time And
A queue estimation system, comprising: a queue estimation unit that calculates a queue length based on the calculated average service time and average arrival time. An average service time calculation step of calculating, as an average service time, a time from a service start time to a service end time included in the provided service log by the average service time calculation unit;
The arrival time calculation unit divides the service time calculated based on the service start time and the service end time included in the log by the total number of the services processed during the service time, thereby obtaining an average arrival time. An arrival time calculating step for calculating
An average waiting time calculating step of calculating an average waiting time based on the calculated average service time and the average arrival time by a queue estimation unit. An average service time calculation step of calculating, as an average service time, a time from a service start time to a service end time included in the provided service log by the average service time calculation unit;
The arrival time calculation unit divides the service time calculated based on the service start time and the service end time included in the log by the total number of the services processed during the service time, thereby obtaining an average arrival time. An arrival time calculating step for calculating
A queue length calculation method comprising: a queue length calculation step of calculating a queue length based on the calculated average service time and the average arrival time by a queue estimation unit. An average service time calculation step of calculating, as an average service time, a time from a service start time to a service end time included in the provided service log by the average service time calculation unit;
The arrival time calculation unit divides the service time calculated based on the service start time and the service end time included in the log by the total number of the services processed during the service time, thereby obtaining an average arrival time. An arrival time calculating step for calculating
An average waiting time calculation step of causing a computer of the queue estimation system to execute an average waiting time calculating step of calculating an average waiting time based on the calculated average service time and the average arrival time by the queue estimation unit. program. An average service time calculation step of calculating, as an average service time, a time from a service start time to a service end time included in the provided service log by the average service time calculation unit;
The arrival time calculation unit divides the service time calculated based on the service start time and the service end time included in the log by the total number of the services processed during the service time, thereby obtaining an average arrival time. An arrival time calculating step for calculating
A queue length calculating step of calculating a queue length based on the calculated average service time and the average arrival time by a queue estimation unit; Length calculation program.

Images