JPH10247214A - Fund predicting method and cash managing system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To calculate the replenishing amount or the recovering amount of cash with respect to an automatic teller machine corresponding to a predicted fund amount necessitated in the case of operating the automatic teller machine from the present time to a designated date based on cash information. SOLUTION: An information collecting part 71 collects each cash information of each ATM 2 within an office A and ATM 22 within an unmanned shot B. An inference engine preparing and updating part 74 prepares an inference engine for predicting a fund amount based on group characteristic and knowledge characteristic and a predicting part 75 predicts the fund amount each ATM necessitates based on the inference engine prepared by the part 74. Based on this fund amount, the replenishing amount or recovering amount of cash is calculated and a present amount transition graph and a replenishing or recovering list are prepared. In addition an operating and displaying part 76 sets a spare amount and a minimum amount as margins and operates others to prepare the present amount transition graph and the replenishing or recovering list.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION The present invention relates to a financial institution such as a bank, for example, based on cash information obtained from a plurality of automatic transaction devices for automatically performing cash deposit / withdrawal transactions.
The present invention relates to a fund prediction method for predicting a required fund amount of each automatic transaction device, and a cash management system using the fund prediction method.

[0002]

2. Description of the Related Art In recent years, many automatic transaction devices (hereinafter, sometimes abbreviated as ATM) such as automatic teller machines have been installed in a large number of mechanized corners of financial institutions such as banks, and have been provided with magnetic cards. It is possible to automatically carry out transactions such as withdrawal (withdrawal) or deposit (deposit) of cash (banknotes and coins) by the user himself / herself using a transaction medium such as a credit card or a magnetic passbook. Has been operating for an extended period of time after the closing of the business, and there are also unmanned stores where only ATMs are installed and unmanned operations installed inside stores such as department stores and supermarkets. It is expected that the number of installations will increase more and more in the future.

[0003] As described above, automatic transaction devices have become widespread, and it has become difficult to manage unattended and holidays. In addition, as the number of installations increases, the amount of cash set in the automatic transaction apparatus has become enormous.

[0004] However, when a plurality of automatic transaction devices are installed side by side, the frequency of use of the automatic transaction devices differs depending on the arrangement. Generally, the utilization rate of an automatic transaction apparatus far from the entrance is low, and the utilization rate is higher as the automatic transaction apparatus is located closer to the entrance.
In such a case, only some of the automatic transaction devices have a small cash balance, and the other automatic transaction devices have a relatively large cash balance. As a result, some automatic transaction devices have no cash, and other automatic transaction devices may have a large amount of cash remaining.

[0005] Therefore, a large amount of cash is set in order to prevent the cash in the automatic transaction apparatus from being high, and the optimum amount is determined based on the experience of a staff member. did not become. In addition, when the automatic transaction apparatus has no cash, it is simply reported as "out of cash", so the required replenishment amount was not known. In addition, since the replenishment of cash must be performed at different times for each automatic transaction apparatus, the efficiency of the business related to the replenishment of cash has been poor.

[0006] Further, the ratio of denominations to be dispensed depends on the store,
Because it changes depending on the day of the week, the end of the month, the beginning of the month, etc., it is difficult for agents to consider these factors and make efficient fund forecasts,
If the clerk makes a mistake, the ratio of the denominations is wasted, so the frequency of cash replenishment increases, and efficient fund management becomes impossible.

Therefore, a fund forecasting method for predicting a necessary fund amount using knowledge such as date, day of the week, and weather has recently been considered.
No. 182,054.

The fund forecasting method disclosed in Japanese Patent Application Laid-Open No. HEI 5-182054 uses the correction signal to convert the time-series data of the forecasted money amount indicating the change of the money amount in the money handling system into the time-series data of the actual actual money amount of the day. Based on the corrected estimated fund amount time-series data, new estimated fund amount time-series data is set as the new estimated fund amount time-series data. When the threshold is exceeded, the actual amount of money time-series data is predicted to change in the future based on the estimated amount of money time-series data, and the amount of money to be collected and replenished is determined. The money in the processing system is prevented from becoming too large and too small, and sudden collection / replenishment of money is prevented.

[0009]

However, according to the fund forecasting method disclosed in Japanese Patent Laid-Open No. Hei 5-182054,
The range of prediction is fixed for a certain period, and the range of prediction cannot be arbitrarily changed. For this reason, for example, there is a problem that it is difficult to match the operation of each branch of the financial institution.

Further, for example, when the present invention is applied to an automatic transaction apparatus for performing cash deposit / withdrawal transactions, it can be applied only to a single automatic transaction apparatus. Therefore, for example, when a plurality of automatic transaction devices are installed in a mechanized corner of a financial institution to manage cash of the plurality of automatic transaction devices, the required amount of funds of the plurality of automatic transaction devices as a whole is There is a problem that it cannot be predicted.

Accordingly, the present invention provides a method of estimating a required amount of funds by using a graph, for example, and thereby changing the required amount of funds arbitrarily. It is an object to provide a method and a cash management system.

It is another object of the present invention to provide a cash prediction method and a cash management system which can easily predict the required amount of funds of a plurality of automatic transaction apparatuses as a whole.

[0013]

The fund forecasting method of the present invention is as follows.

[0014] Further, the fund prediction method of the present invention provides an automatic transaction apparatus for performing cash transactions, in a fund prediction method for predicting the amount of funds required for the automatic transaction apparatus, an arbitrary method for performing fund prediction. Specify the date and time, based on the cash information of the automatic transaction device, predict the amount of funds required to operate the automatic transaction device from the current time until the specified date and time, according to the predicted amount of funds, The amount of replenishment or collection of cash for the automatic transaction device is calculated, and the calculated amount of replenishment or collection is displayed.

Further, according to the fund forecasting method of the present invention, in the fund forecasting method for predicting the amount of funds required for the automatic transaction device for cash transactions, various conditions for performing the fund forecasting are provided. As the current date and time, the business day of the financial institution where the automatic transaction device is installed, the operating time zone of the automatic transaction device, the installation status of the automatic transaction device, the margin amount for the fund amount, and the like, and perform the fund forecasting For any given date and time, based on the various conditions set and the cash information of the automatic transaction device, predict the amount of funds required to operate the automatic transaction device from the current time until the specified date and time Then, in accordance with the predicted amount of funds, the amount of replenishment or collection of cash for the automatic transaction apparatus is calculated, and the calculated amount of replenishment or collection is displayed. The features.

[0016] Further, the fund prediction method of the present invention provides an automatic transaction apparatus for performing cash transactions, in a fund prediction method for predicting the amount of funds required for the automatic transaction apparatus, an arbitrary method for performing fund prediction. A date and time is specified, and based on the cash information of the automatic transaction device, the amount of funds required to operate the automatic transaction device from the current time to the specified date and time is predicted, and the automatic amount is calculated based on the predicted amount of funds. A cash amount transition graph of the transaction device is created, and based on the created cash amount transition graph, a replenishment amount or a collection amount of cash for the automatic transaction device is calculated, and the created money amount transition graph is calculated. And the calculated amount of replenishment or collection of cash is displayed.

Further, according to the fund prediction method of the present invention, in a fund prediction method for predicting a required amount of money of each of a plurality of automatic transaction devices that perform cash transactions, the fund prediction is performed. Specify an arbitrary date and time to perform, based on the cash information of the automatic transaction device, predict the amount of funds required to operate each of the plurality of automatic transaction devices from the current time to the specified date and time, Based on the predicted amount of funds, a list showing the amount of replenishment or collection of cash of each of the plurality of automatic transaction devices is created, and the created list of the amount of replenishment or collection is displayed. .

Further, the cash management system of the present invention is a cash management system which predicts and manages a required amount of money for each of a plurality of automatic transaction devices which perform cash transactions. Management means for managing cash information such as cash balance of each of the plurality of automatic transaction devices, designation means for designating an arbitrary date and time for performing fund prediction, and the plurality of management means managed by the management means Prediction means for predicting the amount of funds required when operating each of the plurality of automatic transaction apparatuses from the current time to the date and time designated by the designation means, based on the cash information of each of the automatic transaction apparatuses, Calculating means for calculating a replenishing amount or a collecting amount of cash of each of the plurality of automatic transaction devices in accordance with the amount of funds predicted in the above, and cash calculated by the calculating means It is provided with display means for displaying the replenishing amount or recovery amount.

Further, the cash management system according to the present invention is a cash management system which predicts and manages a required amount of money for each of a plurality of automatic transaction devices which perform cash transactions. Management means for managing cash information such as the cash value of each of the plurality of automatic transaction devices, and the current date and time as various conditions for performing fund prediction, the business day of the financial institution where the automatic transaction device is installed, Condition setting means for setting the operating time zone of the automatic transaction apparatus, the installation status of the automatic transaction apparatus, a margin amount for the amount of funds, etc .; specifying means for specifying an arbitrary date and time for performing fund forecasting; Based on the various conditions set in the above and the cash information of each of the plurality of automatic transaction devices managed by the management means, from the present time to the designation means A prediction unit for predicting a required amount of money when each of the plurality of automatic transaction devices is operated until a designated date and time, and a cash of each of the plurality of automatic transaction devices according to the amount of fund predicted by the prediction unit. And a display means for displaying the amount of cash replenishment or collection calculated by the calculation means.

Further, the cash management system of the present invention is a cash management system which predicts and manages a required amount of money for each of a plurality of automatic transaction devices which perform cash transactions. Management means for managing cash information such as cash balance of each of the plurality of automatic transaction devices, designation means for designating an arbitrary date and time for performing fund prediction, and the plurality of management means managed by the management means Prediction means for predicting the amount of funds required when operating each of the plurality of automatic transaction apparatuses from the current time to the date and time designated by the designation means, based on the cash information of each of the automatic transaction apparatuses, Based on the amount of funds predicted in the above, processing means for creating a cash amount transition graph of the automatic transaction device, and the cash movement graph created by this processing means Calculating means for calculating the amount of replenishment or collection of cash for the automatic transaction device, and displaying the graph of the amount of cash in the cash created by the processing means, and calculating the amount of cash calculated by the calculating means. Display means for displaying the replenishment amount or the collection amount.

Further, the cash management system of the present invention is a cash management system which predicts and manages a required amount of money for each of a plurality of automatic transaction devices which perform cash transactions. Management means for managing cash information such as cash balance of each of the plurality of automatic transaction devices, designation means for designating an arbitrary date and time for performing fund prediction, and the plurality of management means managed by the management means Prediction means for predicting the amount of funds required when operating each of the plurality of automatic transaction apparatuses from the current time to the date and time designated by the designation means, based on the cash information of each of the automatic transaction apparatuses, Calculating means for calculating the amount of replenishment or collection of cash of each of the automatic transaction apparatuses according to the amount of funds predicted in the above; and the automatic calculation means calculated by the calculating means. Means for creating a list showing the amount of cash replenishment or collection of each cashing device, and a list showing the amount of cash replenishment or collection of each automatic transaction device created by this creation means. Display means.

A cash management system according to the present invention is a cash management system which predicts and manages a required amount of money for each of a plurality of automatic transaction devices that perform cash transactions. Management means for managing cash information such as cash balance of each of the plurality of automatic transaction devices, designation means for designating an arbitrary date and time for performing fund prediction, and the plurality of management means managed by the management means Prediction means for predicting the amount of funds required when operating each of the plurality of automatic transaction apparatuses from the current time to the date and time designated by the designation means, based on the cash information of each of the automatic transaction apparatuses, Calculating means for calculating a replenishing amount or a collecting amount of cash of each of the plurality of automatic transaction devices in accordance with the amount of funds predicted in the above, and cash calculated by the calculating means Display means for displaying a replenishment amount or a collection amount, and a cash amount transition data creating means for creating cash amount transition data for each of the plurality of automatic transaction devices based on the fund amount predicted by the prediction means, The cash flow transition data created by the cash flow data creation means is monitored, and when the cash flow transition is abnormally different from the cash transition predicted by the prediction means, a person in charge is notified of the fact. And a prediction monitoring means for giving a warning.

Further, the cash management system according to the present invention is a cash management system for managing an operation state including a cash amount of a plurality of automatic transaction devices installed in a store of a financial institution. Has a group management function with a carrier that stores cash to be replenished or collected for a plurality of automatic transaction devices installed in a store and moves between the plurality of automatic transaction devices, In a store provided with, the group management function is used to manage cash of a plurality of automatic transaction devices, and the management of managing cash information such as cash amount of each of the plurality of automatic transaction devices. Means, designating means for designating an arbitrary date and time for performing fund prediction, setting means for previously setting a first reserve amount as a margin for performing fund prediction, Based on the cash information of each of the plurality of automatic transaction devices managed by and the reserve amount set by the setting unit, each of the plurality of automatic transaction devices from the present time to the date and time designated by the designation unit And a fund estimating unit for estimating the amount of cash required for operating the plurality of automatic transaction devices according to the estimated amount of cash, and calculating the amount of replenishment or collection of cash for each of the plurality of automatic transaction devices. When calculating the amount of cash replenishment or collection for each of the plurality of automatic transaction devices, the fund prediction means uses the first reserve amount set by the setting means to refill or collect cash. Calculating means for calculating the amount of cash replenishment or collection as a whole of the plurality of automated teller machines at a store where the carrier is not installed, The first set Te 1
A second reserve amount obtained by adding the number of automatic transaction devices installed to the reserve amount, and calculating a replenishment amount or a recovery amount of cash using the second reserve amount Means and a third reserve amount smaller than the second reserve amount when calculating a replenishment amount or a collection amount of cash as a whole of the plurality of automatic transaction devices at a store where the carrier is installed. And a third calculating means for calculating the amount of replenishment or collection of cash using the third reserve amount.

[0024]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a conceptual diagram of a cash management system in a bank according to the present invention. In FIG. 1, A is a branch office (store) of a bank, B is an unmanned store of a bank, C is a calculation center, D is a centralized center for centrally managing the back office work of each branch office A and the operation of each ATM. Is a security company that manages the ATM of the unmanned store B, and F is a business partner such as a department store or a supermarket.

In the store A, an in-store LAN (local
Area network) 1 is provided. LA in the store
N1 includes a plurality of automatic transaction devices (automatic teller machines, cash dispensers, bookkeeping machines, balance inquiry machines, etc.) 2 installed at mechanized corners in the store, etc. Robot controller 4 that wirelessly controls a robot (cash transfer device) 3 as a carrier for group recycling management that automatically transfers funds, a counter terminal 5, a cash processor 6, a calculation center modem 7, an unmanned store Management server 13 for managing all cash in the own store and the unmanned store B under management, comprising a modem 8, a public relations terminal 9, a sales office server 10, a customer database 11 and a cash database 12, and a modem 14 for a centralized center. Etc. are connected.

An in-store LAN 21 is provided in the unmanned store B. The in-store LAN 21 has a plurality of automatic transaction devices (automatic teller machines, cash dispensers, etc.) installed in the store. ) 22,..., A business office modem 23 and the like. The office modem 23 is connected to the communication line 31.
Is connected to the unmanned store modem 8 in the sales office A through
Communication is performed between the two.

In the calculation center C, the customer database 4
1 is provided. The host computer 42 is connected to the modem 7 for the calculation center in the sales office A via the communication line 32, and communicates between them.

In the centralized center D, the in-center LAN 51
Is provided. The in-center LAN 51 has a business partner database 52 and a cash database 53, a fund management server 54 for managing all cash in the own center and the office A under management, a plurality of cash processors 55,. A modem 56, a modem 57 for a calculation center, a modem 58 for a sales office, a modem 59 for a business partner, and the like are connected.

The modem 57 for the calculation center is connected to the communication line 33.
Is connected to the host computer 42 in the calculation center C via the PC and communication between them is performed. The branch office modem 58 is connected to the centralized center modem 15 in the branch office A via the communication line 34, and performs communication between them. The business partner modem 59 is connected to the cash management server of the business partner F via the communication line 35 and communicates with each other.

In the security company E, an in-company LAN 61 is provided. The in-company LAN 61 has unmanned store information for viewing the ATM cash information of the unmanned store B under management via the centralized center D. A terminal 62, a central center modem 63, and the like are connected. Centralized modem 63
Is connected to the security company modem 56 in the centralized center D via the communication line 36, and communicates between them.

FIG. 2 schematically shows a configuration of a portion for performing group recycling management using the robot 3. In FIG. 2, a cash management server 13 as an operation management device has a plurality of, for example, 4
Automatic transaction devices (hereinafter sometimes abbreviated as ATM) 2a, 2b, 2c, 2d. Cash information such as the operating state (for example, a failure occurrence state) and the remaining amount of cash (banknotes and coins) is reported from the ATMs 2a to 2d to the cash management server 13 via the in-store LAN 1 at any time. ing.

The cash management server 13 has a monitor 1
The robot controller 4 is connected to the robot 3 via the controller 5, and the robot 3 is wirelessly controlled by the robot controller 4. Under the control of the robot controller 4, the robot 3 automatically travels between the ATMs 2a to 2d while
Replenishment and collection of cash are performed for M2a to 2d.

FIG. 3 more specifically shows a configuration of a portion for performing group recycling management using the robot 3 shown in FIG. In FIG. 3, ATMs 2a to 2d are installed side by side on a floor 101 of a financial institution with their rear surfaces aligned in parallel. The ATM 2a has a substantially rectangular box-shaped main body 102, and a substantially L-shaped customer service panel 103 facing the user is provided on the front surface of the main body 102. The customer service panel 103 is provided with an operation display unit also serving as a touch panel, a bill receiving / dispensing port opened / closed by a door, a coin receiving / dispensing port, a card insertion port, a passbook insertion port (all not shown), and the like. Since these are already well-known technologies, further description will be omitted.

In the main body 102, there are provided a bill receiving / dispensing unit 104 for depositing / dispensing bills to a user, a coin receiving / dispensing unit (not shown) for depositing / dispensing coins, and the like. These banknote deposit / withdrawal units 104
The coin depositing and dispensing unit constitutes a cash storage unit in the present invention. Further, on the rear surface of the main body 102, a carry-in port 106 facing the cassette loading section 105 of the bill receiving / dispensing unit 104 is formed. The carry-in entrance 106 is opened and closed by, for example, a child door 107 that opens up and down.

The other ATMs 2b, 2c and 2d have the same configuration as the above-mentioned ATM 2a.

On the other hand, the robot 3 receives a command wirelessly transmitted from the robot controller 4 via the antenna 108 via the antenna 109, and thereby, for example, along the rear surfaces of the ATMs 2a to 2d on the floor surface 101. It is provided so as to be able to travel on two installed rails 110.

The robot 3 is provided with two drive wheels 111, and the two drive wheels 111
It is possible to run on 10 above. In addition, the robot 3
One loading cassette 112 is always detachably held. The loading cassette 112 is configured to store therein a large number of banknotes in a stacked state. Robot 3
In accordance with a command from the robot controller 4, the vehicle self-propelled to a position facing a predetermined ATM,
The banknotes are transferred between the cassette loading unit 105 and the loading cassette 112 via a carry-in port 106 provided on the rear surface.

Next, a fund prediction method according to the present invention will be described.

First, an outline of the fund prediction will be described.
FIG. 4 shows a place where cash is present in a sales office. The hatched portion in the figure indicates the amount of cash. The cash management server manages ATMs, teller counters, unpaid accounting systems (cash handling machines), cash receipts, large safes, and external liaisons to manage cash balance and deposit / withdrawal information (cash information). It can make a forecast of a large fund.

More specifically, for example, when looking at an ATM, there is generally a variation in the amount of money depending on each car. FIG.
In the graph, the horizontal axis represents the date, and the vertical axis represents the required amount of funds, and represents, for example, the cumulative required amount of funds for 17 ATMs in a certain branch. It is assumed that the 17 ATMs are arranged, for example, as shown in FIG. 6, and three-digit numerical values (302 to 310, 902 to 909) in the figure indicate the ATM number.

As can be seen from FIGS. 5 and 6, there is a variation in the required fund amount depending on the car. In particular, near the entrance and far away, A near the frequently used entrance
TMs tend to run out of money sooner, while on the other hand there is a tendency for the amount of cash to not readily decrease far from the entrance. Using this effectively, ATMs with similar trends
Into groups. In the example of FIG.
Are divided into seven groups. This is called “group characteristics”.

Further, it has been found that by collecting data to some extent, elements such as the day of the week and the date tend to have more money. For example, FIG. 7 shows a tree diagram of knowledge based on days of the week and dates in the data example shown in FIG. First, look at the day of the week of the corresponding forecast. Further, it is determined whether or not it is the 25th of the payday, and further, whether or not it is the day before the holiday, and finally, the necessary amount of funds is calculated in consideration of the above group characteristics. The required amount of money is, for example, 100 units, so all fractions are rounded up. This is called "knowledge characteristics".

By combining these two "group characteristics" and "knowledge characteristics", it is possible to predict the optimal amount of funds for each ATM.

Here, the group characteristics will be described. FIG. 5 shows, for example, the transition of the required amount of funds for one month. This required amount of money is obtained from the difference between the sum of the deposit and the sum of the withdrawal in the respective transactions of deposit, withdrawal and cash transfer. For example, FIG. 9 shows the transition of the accumulated deposit amount for one month, and FIG. 10 shows the transition of the accumulated deposit amount for one month. From these two graphs, the graph of FIG. 5 (the required amount of funds) is obtained, and those having similar tendencies are collected and grouped.

For example, the grouping will be described with reference to FIG. The required amount of funds is divided into 1000 units.
If it exceeds the range, it belongs to the next group. In addition, the required amount of money varies because the transaction amount differs every month, and accordingly, the group attribute of each ATM also changes. Therefore, the number of groups changes every month.

Next, the effectiveness of the grouping will be described with reference to FIG. For example, ATM1 in FIG.
Looking at the fund management of ATM2, (a) A failure occurred during the operation of ATM1 the day before,
It stopped working, it was used less frequently than usual, and the amount of money (funds) did not decrease much. In comparison, ATM2 has A
Because it was not used in TM1, the frequency of use of ATM2 was higher than usual, and the amount of money was lower than expected.

(B) When making a prediction on the current day, if grouping is not performed, an instruction is issued to ATM1 indicating that replenishment is not required, and an instruction is issued to ATM2 indicating that an extremely large amount of replenishment is required.

(C) As a result of being instructed and replenished in this way, ATM1 is used more than expected and runs out of funds, and ATM2 is used less than expected and becomes overfunded.

(D) On the other hand, when making predictions on the day, by performing grouping, it is necessary to slightly replenish ATM1 and instruct ATM2 to replenish it appropriately.

(E) Thereby, the operation can be performed without any problem on the day.

As described above, the merit of the grouping is that the prediction accuracy can be improved without being affected by noise (such as a machine stoppage due to a failure or a special date) in a situation where the amount of data at the initial stage of system introduction is small. Changes in the external environment (A
Expansion of TM, construction of residential complex, etc.). (Because there is a case where it is not possible to judge whether or not each ATM is a noise.) Next, the fund prediction method according to the present invention will be specifically described.

FIG. 13 shows the cash management server 1 in the branch A.
3 shows a functional configuration diagram of a portion for performing fund prediction, and includes an information collection section 71, a group characteristic creation / update section 7
2. It is composed of a knowledge characteristic creation / update section 73, an inference engine creation / update section 74, a prediction section 75, and an operation / display section 76.

The information collecting unit 71 is provided for each ATM in the branch A.
2 and each cash information of each ATM 22 in the unmanned store B is collected. The group characteristic creating / updating unit 72 determines each ATM based on the cash information collected by the information collecting unit 71.
Of the total deposit and withdrawal of the A
As described above, each ATM is grouped to create a group characteristic based on the accumulated deposit amount and the accumulated deposit amount of the TM, and the created group characteristic is updated.

The knowledge characteristic creating / updating unit 73 creates a knowledge characteristic in which a necessary amount of money is generated corresponding to elements of the day of the week and date, and updates the created knowledge characteristic.
The inference engine creating / updating unit 74 performs group property creating /
Based on the group characteristics created by the update unit 72 and the knowledge characteristics created by the knowledge characteristic creation / update unit 73,
For each ATM, create an inference engine to predict the required amount of money for each day of the week and date,
The created inference engine is updated.

Based on the inference engine created by the inference engine creating / updating unit 74, the prediction unit 75
And the amount of cash replenishment or collection based on this estimated amount of cash,
This is used to create a later-described money amount transition graph and a list of replenishment or collection. As shown in FIG.
1, a graph creation unit 82 and a list creation unit 83.

The operation / display unit 76 is used for setting a reserve amount and a minimum amount as a margin, performing other operations, and displaying a remaining amount transition graph and a list of replenishment or collection described later. And a display unit such as a CRT display.

In such a configuration, an outline of the fund prediction process will be described with reference to a flowchart shown in FIG. When an instruction for fund prediction is given, first, in step S1, the information collecting unit 71 collects cash information of each ATM 2 in the sales office A and each ATM 22 in the unmanned store B.

Next, in step S2, the group characteristic creating / updating unit 72 checks each A collected in step S1.
Based on the cash information of the TM, the total deposit and the total withdrawal of each ATM are obtained, and based on the obtained total of the deposit and the total withdrawal of each ATM, as described above, the ATMs with similar tendencies are determined. A group characteristic is created by grouping, and the created group characteristic is updated.

Next, in step S3, the knowledge characteristic creating / updating unit 73 creates a knowledge characteristic that generates a necessary amount of money corresponding to the day of the week and date elements, and updates the created knowledge characteristic. Perform

Next, in step S4, the inference engine creating / updating unit 74 determines the day of the week, the day of the week, and the like for each ATM based on the group characteristics created in step S2 and the knowledge characteristics created in step S3. Create an inference engine for estimating the required amount of funds for each date, and update the created inference engine.

Next, in step S5, the prediction unit 75
Predicts the required amount of money for each ATM based on the inference engine created in step S4, and replenishes the amount of cash based on the predicted amount of money and the spare amount set in the operation / display unit 76. Alternatively, the collection amount is calculated.

Hereinafter, each step in FIG. 12 will be described in detail.

First, the creation of the group characteristics in step S2
The update process will be described with reference to the flowchart shown in FIG. First, in step S11, it is determined whether or not the operation day is the last day of the month. If it is not the last day of the month, the process proceeds to step S12, where the required amount of money for each ATM in units of one day is paid and received for one day of transactions. Calculated from

In step S11, if the operating day is the last day of the month, the operation proceeds to step S13, where the required amount of money for each ATM in units of one month is obtained, and the operation proceeds to step S14. In step S14, according to the above-described grouping rule, as shown in FIG. 17, each ATM is grouped to create a group characteristic.

Next, in step S15, it is determined whether or not there is the last month's group characteristics. If not, the process is terminated. If there is, the process proceeds to step S16, where the last month's group characteristics are updated to the group characteristics created this time. Then, the process ends.

Next, the knowledge characteristic creation / update processing in step S3 will be described with reference to the flowchart shown in FIG. First, in step S21, it is determined whether or not the operation day is the last day of the month. If it is not the last day of the month, the process proceeds to step S22 to input what week of the week and what day of the week.

Next, in step S23, it is determined whether or not the operation day is a pay day (for example, 25 days). If it is a pay day, the process proceeds to step S24, and the operation day is Enter the day of the week and payday.

In step S23, if the operation day is not a payday, the process proceeds to step S25, and it is determined whether the operation day is a day before a holiday. If not, the process returns to step S21, and the same operation as described above is performed. repeat. In step S25, if the operation day is a day before a holiday, the process proceeds to step S26. The operation day is input on what day of the week and the day before the holiday, and the process returns to step S21. repeat.

If the operation day is the last day of the month in step S21, the process proceeds to step S27. In step S27, as shown in FIG. 19, knowledge characteristics of each day in one month are created, and the process proceeds to step S28. In step S28,
It is determined whether or not there is the knowledge characteristic of last month, and if not, the process is terminated. If there is, the process proceeds to step S29, where the knowledge characteristic of last month is updated to the knowledge characteristic created this time, and the process is terminated.

Next, in step S4, an inference engine is created.
The update process will be described with reference to the flowchart shown in FIG. First, in step S31, it is determined whether or not the operation day is the last day of the month. If not, the process waits until the last day of the month.

If the operation day is the last day of the month in step S31, the process proceeds to step S32. In step S32, as shown in FIG. 21, an inference engine is created from the previously created group characteristics and knowledge characteristics.
Proceed to. In step S33, it is determined whether or not there is an inference engine of last month. If there is no inference engine, the process is terminated. If there is, the process proceeds to step S34 to update the inference engine of last month with the inference engine created this time, and the process ends. .

Next, the fund prediction processing in step S5 will be described with reference to the flowchart shown in FIG.
First, it is determined to which of the previously created groups the ATM to be funded belongs (S41), and the determination result is stored in a memory (not shown) (S42).

Next, each content of the previously created knowledge characteristic will be examined. That is, first, the day of the week is determined (S43), and the determination result is stored in a memory (not shown) (S44, S45, SA46). If the result of determination is Monday to Friday, it is determined whether or not it is a payday (S47, S48), and the result of the determination is stored in a memory (not shown) (S49, S50).

Further, if the result of determination is Monday to Thursday, it is determined whether or not it is a day before a holiday (S51), and the result of determination is stored in a memory (not shown) (S52).

The fund prediction is performed based on these stored data, and the data to be referred to changes depending on whether or not the data has been stored, for example, for one year or more. That is, it is determined whether or not the data has been stored for one year or more (S53). If there is data for one year or more, by referring to the average data of the data one month ago and the data one year ago (S53). (S54), the required funding amount is calculated (S55). If there is no data for one year or more, by referring to the data one month ago (S5
6), calculate the required amount of funds (S55).

Next, after adding a reserve amount as a predetermined margin to the required amount of money calculated as described above (S57), the final replenishment amount is determined (S58).

As described above, various data necessary for fund prediction are sequentially stored in the memory as shown in FIG. 23, and fund prediction is performed based on the stored data.

Here, the setting of the margin amount and the transition of the money amount in step S57 of FIG. 22 will be described with reference to FIG. In FIG. 24, a is the current amount of money, b
Indicates the required amount of funds, c indicates the amount of instructions for replenishment and collection to the staff, and d indicates the actual changes in funds.

Now, the replenishment and collection of cash is carried out first in the morning,
It is assumed that the margin amount is 200 sheets. In the morning of the day, the amount of money was 600, and the forecast for that day (the required amount of funds) was 8
Assuming that the number is 00, the amount of margin is added to the required amount of funds (8
00 + 200), the current amount of money is subtracted to determine the replenishment amount (800 + 200−600 = 400). Similarly, the replenishment amount is determined on the second and third days.

Next, a description will be given of fund estimation until data is accumulated for one month. The group characteristics described so far cannot be grouped unless data accumulates for one month. Therefore, since it is not possible to make a fund forecast, one month after the introduction of the cash management system, the fund forecast is simply performed by the method described below.

For example, a case where the present cash management system is introduced on Monday will be described with reference to a flowchart shown in FIG. 25 and FIG. When the operation on the first day is completed (S61), the process proceeds to step S62, and the data on the first day is acquired as a reference amount. Next, step S63
In, judge how long the operation days, from the second day to 7
In the case of the day, the process proceeds to step S64, and the fund prediction is performed based on the reference amount acquired in step S62.

Next, in step S65, the data of the previous day is obtained as a reference amount, and the flow advances to step S66. In step S66, it is determined whether or not it is day 7, and if it is not day 7, the process returns to step S63 to repeat the above operation.
If it is the day, the process proceeds to step S67. In step S67, one week's performance data is acquired, and the process returns to step S63 to repeat the above operation.

In step S63, on the eighth day,
Proceeding to step S68, fund prediction is performed based on the previously acquired one-week performance data. Next, in step S69, it is determined whether or not it is day 14, day 21 or day 28,
If so, the process proceeds to step S70. In step S70, the result data of the previous week is acquired, and in step S71
Proceed to. If it is not the 14th, 21st, or 28th day in step S69, the process proceeds to step S71.

In step S71, it is determined whether or not one month is over (the 32nd day). If one month is not over, the process returns to step S68 and the above operation is repeated. Step S71
If one month is over, the process proceeds to the ordinary fund prediction processing based on the group characteristics and the knowledge characteristics described above.

As described above, on the first day after the introduction of the cash management system, data on the first day is obtained as a reference amount. On the second day, the fund is simply predicted based on the data on the first day. The third day is the second day data, the fourth day is the third day data, the fifth day is the fourth day data, the sixth day is the fifth day data, and the seventh day is the sixth day. Funds are simply forecasted based on eye data.

At the end of the operation on the seventh day,
One-week performance data is created, and based on that data, a second-round fund forecast is made. Similarly, in the third and fourth laps, fund prediction is performed based on the actual data of the previous week. Then, at the end of one month, normal fund prediction is performed based on the group characteristics and the knowledge characteristics described above.

Next, the flow from the introduction of this system to daily operation will be described with reference to the flowchart shown in FIG. First, in step S81, various system settings (condition settings) are performed. Here, the system settings include “current date and time setting”, “financial institution business day setting”, “ATM operating time zone setting”, “ATM installation status setting”, “cash replenishment and collection instruction unit”. "Setting" and "margin setting", which will be described later.

Next, after the system is set, the operation is performed by simple prediction with reference to the above-mentioned daily reference amount until one month has passed since the system was introduced (S82, S8).
3). Then, one month after the introduction of the system, the forecasting method is automatically switched (S84), and ordinary fund forecasting is performed by inference based on the above-mentioned pattern and knowledge (S85). The above pattern is automatically updated, for example, every month.

If an ATM is added during operation in this manner (S86), the flow returns to step S81 to change the setting of the ATM installation state and start again from the simple prediction. For other system changes, change the initial settings of necessary items.

Here, the system setting in step S81 will be described in detail. The display screen example used in the following description is displayed on the display unit of the operation / display unit 76.

First, the "current date and time setting" will be described with reference to FIG.
This will be described using the display screen example shown in FIG. The current date and time is always set because it is the origin of the fund forecast. Before the change, the currently set time is displayed, and after the change, the correct date and time are input. That is, by selecting “current date and time setting” on the user setting screen displayed on the operation / display unit 76, the current date and time setting screen of FIG. 28 is displayed. When this screen is displayed, the cursor is placed on the changed area.

Here, when it is desired to change the date and time, the date and time of the post-change area are re-entered to change the date and time. When changing the date and time, the user presses a setting button F1 displayed on the screen to reset the date and time. After the date and time is changed, if the return button F2 displayed on the screen is pressed before the setting button F1 is pressed, the date and time are not changed and the screen returns to the user setting screen.

Next, "setting of business days of financial institution" will be described with reference to a display screen example shown in FIG. For example, when forecasting on Friday, if Saturday and Sunday are set as holidays, the forecast can be made up to Monday. Therefore, the prediction range can be managed by the system.

That is, by selecting “window business day setting” on the system setting screen displayed on the operation / display section 76, the window business day setting screen of FIG. 29 is displayed.
When this screen is displayed, the cursor is placed on today's area. Business days are displayed in blue, and holidays are displayed in red.

By moving the cursor to the date to be set and pressing the space key provided on the keyboard of the operation / display unit 76, the business day and the holiday are switched, and the setting can be made. When all settings have been completed, a setting window F1 displayed on the screen is pressed to display a mini window for selecting whether the current setting is only this year or all. In this mini window, if the return button F2 displayed on the screen is pressed before selecting only this year or all, the contents set so far are cancelled, and the screen returns to the system setting screen. I have.

[0097] Incidentally, when moving to the contents of the previous month or the contents of the next month, it is possible to press the previous month button F3 or the next month button F4 displayed on the screen.

Next, "ATM operating time zone setting" will be described with reference to a display screen example shown in FIG. ATM
By setting the operating time zone of the inference engine, it is possible to know from what time and from what time the inference engine has used it. The operating time zone of the ATM is set by registering a pattern in advance and setting the pattern on each day.

That is, by selecting the monthly setting of “ATM operating time zone setting” on the system setting screen displayed on the operation / display section 76, the ATM operating time zone setting (monthly setting) screen of FIG. 30 is displayed. Is done. When this screen is displayed, the cursor is placed on the first day of the current calendar month. Here, if a pattern is registered for each day of the week, the pattern is displayed.

To change the pattern, move the cursor to the pattern registration area and register the pattern for the ATM operating time zone. In registering a pattern, when a new pattern is registered or the time is changed, a mini window is displayed to prompt the user to determine whether or not the pattern can be updated. The number of registered patterns is six in this example.

When the registration of the pattern is completed, the cursor is placed on the first day of the current month calendar. Here, the pattern is set by inputting the pattern of the day to be changed and pressing the setting button F1 displayed on the screen. Then, when the return button F2 displayed on the screen is pressed, the screen returns to the system setting screen.

Next, "ATM installation status setting" will be described with reference to a display screen example shown in FIG. ATM
When the number of ATMs whose funding is to be predicted increases due to an increase in the number of ATMs, the use state of the ATM changes.

That is, by selecting “ATM installation status setting” on the user setting screen displayed on the operation / display unit 76, the ATM installation status setting screen shown in FIG. 31 is displayed. When this screen is displayed, the cursor is placed on the 01th car. In addition, when the ATM is already installed, it is indicated by oblique lines, and when the ATM is compatible with the group recycling management, it is indicated by crossed oblique lines.

When the number of ATMs for which fund prediction is to be newly increased due to the addition of an ATM, etc., the cursor is positioned on the car and the space key provided on the keyboard of the operation / display unit 76 is pressed to set the ATM. Is set. In this case, when the space key is pressed once, the ATM is installed, and when the space key is pressed twice, it corresponds to the group recycling management.

When the setting is completed, the final decision is made by pressing the setting button F1 displayed on the screen. If the return button F2 displayed on the screen is pressed before the setting button F1 is pressed, the contents input so far are canceled and the screen returns to the user setting screen.

Next, “Cash replenishment / collection instruction unit setting”
Will be described with reference to a display screen example shown in FIG.
It is necessary to set the unit of the replenishment / collection amount finally instructed to the attendant in order to round off fractions. The unit is set.

That is, by selecting “replenishment / collection instruction unit setting” on the user setting screen displayed on the operation / display unit 76, the replenishment / collection instruction unit setting screen shown in FIG. 32 is displayed. When this screen is displayed, a cursor is placed on the minimum bill replenishment / collection area after the change.

For the change setting, move the cursor to the item to be input and input. After inputting the change contents, press the setting button F1 displayed on the screen,
Finalize. If the return button F2 displayed on the screen is pressed before the setting button F1 is pressed, the changed content is canceled and the screen returns to the user setting screen.

Next, “margin setting” will be described with reference to FIG.
This will be described using the display screen example shown in FIG. As previously mentioned,
A margin (spare amount) is added to the estimated necessary amount of funds, and the final instruction amount of replenishment and collection to the staff is made. The margin is set.

That is, by selecting “margin setting” on the manufacturer setting screen displayed on the operation / display section 76, the margin setting screen shown in FIG. 33 is displayed. When this screen is displayed, the cursor is placed on the No. 01 machine, and the calendar displays the month of today's date. Here, when the number is selected and the cursor is set on the calendar, the spare number and the minimum number corresponding to the date are displayed.

To change the setting, a date to be changed is selected by a calendar, and further, by selecting a setting unit, the spare number and the minimum number are input to the changed number. When the input is completed, the setting change is confirmed by pressing the setting button F1 displayed on the screen. If the return button F2 displayed on the screen is pressed before the setting button F1 is pressed, the values changed so far are canceled and the screen returns to the manufacturer setting screen.

Further, at the time of group recycling management, a margin is set by setting the unit area to all the units.

Next, the flow of daily operation of the present system will be described with reference to the flowchart shown in FIG. I want to replenish and collect cash in ATMs on a daily basis.
If you need to know how much
By depressing the operation support key provided on the keyboard of the display unit 76 (S91), the whole bill graph screen as shown in FIG. 35 is displayed on the display unit of the operation / display unit 76 (S92).

Hereinafter, the bill whole graph screen of FIG. 35 will be described. This screen can be used for the amount of funds to be brought out of the ATM when replenishing cash and the amount of money needed for the current receipt, because it is possible to see at a glance how much money is needed and how much money is needed as an ATM corner . The default of this screen display is to move the arrow P1 displayed on the screen along the time axis to display a transition graph of the actual amount of money up to the present according to the current time. In this case, 10,000 yen is displayed in a blue graph, and 1,000 yen is displayed in a red graph.

In the bill whole graph screen shown in FIG. 35, the horizontal axis represents time, but only the operating days and operating hours of the ATM are used as scales. The arrow P1 can be arbitrarily moved along a time axis by a mouse (not shown), and when the arrow P1 is moved, the replenishment number of 10,000 yen and 1,000 yen required by the date and time when the arrow P1 is moved is displayed. A display is also provided.

The time axis can be shifted left and right using the scroll bar P2 displayed on the screen. In this case, the display range is, for example,
The past week and the past week are displayed from the day.

Here, a method of creating a graph on the entire banknote graph screen of FIG. 35 will be described.

As described above, by moving the arrow P1, a prediction graph is displayed. Hereinafter, this will be described with reference to FIG. 36 by taking cash replenishment as an example.
For example, set arrow P1 to 13:00 on 3 (Fri),
It is assumed that prediction is made from 1 (Wed) 13:00 to 3 (Fri) 13:00.

Thus, the required amount of funds for each day is obtained by the above-mentioned inference. This is indicated by a broken line in FIG. As is clear from FIG. 36, 1 (water)
The required amount of funds from 13:00 to 19:00 is a, the required amount of funds on 2 (Thu) is b, and the required amount of funds on 3 (Fri) is c.

Therefore, from 1 (water) 13:00 to 3
(Fri) The amount of funds required by 13:00 is (a + b + c)
Minutes. Therefore, by moving a, b, and c in parallel so as to connect the graphs, a money amount transition graph is created. As a result, d becomes the replenishment amount, and finally a money amount transition graph as shown in FIG. 37 is displayed.

Next, a case of collecting cash will be described with reference to FIG. When the amount of cash is large and it is collected, it is as shown in FIG. For example, it is assumed that the arrow P1 is adjusted to 13:00 on 3 (Fri) and predicted from 13:00 on 1 (Wed) to 13:00 on 3 (Fri).

Thus, the required amount of funds for each day is obtained by the above-mentioned inference. This is indicated by a broken line in FIG. As is clear from FIG. 38, 1 (water)
From 13:00 to 19:00 is -a '
And the required funding amount of 2 (tree) is -b ', and 3
The required amount of money for (Fri) is -c '.

Therefore, from 1 (water) 13:00 to 3
(Fri) The amount of funds required by 13:00 is (-a'-b '
−c ′) minutes. Therefore, by moving -a ', -b', and -c 'in parallel so as to connect the graphs, a money transition graph is created. As a result, d 'becomes the collection amount, and finally a money amount transition graph as shown in FIG. 39 is displayed.

As is clear from the above description, the entire bill graph screen shown in FIG. 35 is a money amount transition graph for a plurality of ATMs. To find the replenishment amount for the entire ATM,
Since the required amount of money for the plurality of ATMs obtained as described above is obtained by adding the reserve amount of the plurality of ATMs as a whole, it is additionally supplemented by the spare amount of the plurality of ATMs. Will be.

Here, the reserve amount changes depending on whether or not the group recycling management as described above is introduced.

When the group recycling management is introduced:
(ATM number x spare amount) x 1/2 When group recycling management is not introduced: (ATM number x spare amount) Thus, if group recycling management is not introduced,
As a result, the reserve amount is the same as that of the individual management (to be described later) of each of the plurality of ATMs.

As described above, the value of the spare amount is input in advance on the margin setting screen. This reserve amount is a necessary value from the past data so that the ATM does not run out of funds.

As described above, the whole banknote graph screen of FIG. 35 is a graph showing a change in the amount of money for a plurality of ATMs. However, when a graph is displayed for each machine (S93), the whole banknote screen of FIG. By selecting a desired car in the car setting area E1 of the graph screen (S94), the money amount transition graph for each car is created and displayed by the same graph creation method as described above (S9).
5).

The replenishment amount of each ATM is determined by adding a reserve amount to the required amount of money of each of the plurality of ATMs determined as described above. (Reserved amount).

If there is a machine that requires replenishment of cash on the whole bill graph screen of FIG. 35, a message to that effect is displayed at the same time. Hereinafter, this will be described.

When displaying the entire banknote graph showing the change in the amount of money of all the units (the entire ATM corner), the ATM corner as a whole has ample funds, but when looking at each unit, the near-end and the end It is possible that a cut has occurred. In such a case, a message to the effect that there is a car running out of funds is displayed simultaneously with the graph, and the staff is instructed to replenish.

For example, as shown in FIG. 35, a message area E2 is provided on the whole bill graph screen, and the message area E2 is displayed. In the figure, for example, it is displayed that Unit 2 is at the end of 10,000 yen and Unit 5 is at the end of 1,000 yen.

If the display area of the message area E2 is not enough, only the message "There is a machine with out of funds" is displayed on the whole banknote graph screen, and the details are as shown in FIG. It may be displayed on another screen.

Here, a description will be given of the fact that the present system grasps the amount of cash. In this system, the amount of money for each denomination is always grasped from each ATM in a state as shown in FIG. That is, as shown in FIG.
Each unit has the amount of money for each denomination as table data, and further stores the latest acquisition time of the amount of money.

[0135] The timing of taking in the cash amount from each ATM is, for example, after the end of the transaction, after turning on the power, after resetting, after a manual stop by the attendant, after loading (replenishment) and after tightening (collecting). .

If the user wants to see a list of replenishment and collection on the whole banknote graph screen of FIG. 35, the replenishment and collection list button F5 displayed on the screen is pressed (S
96), a cash replenishment and collection list is created, and a replenishment and collection list screen as shown in FIG. 42 is displayed (S97).

The replenishment / collection list screen shown in FIG. 42 will be described below. For each unit, when the forecast date and time are changed on the whole bill graph screen, the forecast range is from the present to the changed date and time, and if not changed, the default date and time is the forecast range and the replenishment (collection) amount is Ask and display.
The color is displayed in red for replenishment and blue for recall.

On the replenishment / collection list screen, the prediction range can be changed by moving the cursor to the prediction range area. When the prediction range is changed, the value of the replenishment (collection) amount changes.

If the number of ATMs is large and cannot be displayed on one screen, the display is scrolled. In addition, when the group recycling management is introduced, only the target unit and denomination are displayed to replenish only a specific ATM, and the blank is set thereafter.

The replenishment / recovery list screen of FIG. 42 instructs replenishment of various media. For example, an instruction is issued such that the journal receipt of the ATM is always in full volume in units of one volume. In addition, since the remaining amount of the transfer ticket cannot be accurately grasped, an instruction is given so that the transfer ticket is always full.

In the replenishment / collection list screen shown in FIG. 42, by pressing the return button F2 displayed on the screen, the screen returns to the whole bill graph screen shown in FIG. Also,
On the whole banknote graph screen of FIG. 35 or the replenishment / collection list screen of FIG. 42, the screen display can be ended by pressing the end button F6 displayed on the screen.

Next, the flow of the prediction monitoring of the present system will be described with reference to the flowchart shown in FIG.
During daily operation (S101), contrary to the prediction, for example,
When the withdrawal transaction continues and you are running out of money,
That is, the current balance is compared with a preset reserve amount, and when [current balance ≦ reserve amount] is satisfied (S10).
2), a prediction monitoring screen is displayed on the display unit of the operation / display unit 76 (S103), and a warning sound is issued to urge the attendant to warn. The prediction monitoring screen is a screen similar to the whole banknote graph screen of FIG.

When the attendant presses the alarm release button on the prediction monitoring screen (S104), the warning sound is stopped, and the routine returns to daily operation (S105).

As described above, according to the above-described embodiment, the prediction range of the required amount of funds can be arbitrarily changed using the graph. Therefore, for example, it is possible to flexibly cope with the timing of replenishing or collecting cash at a Friday before Saturday / Sunday or at a branch of a financial institution. Further, it is possible to easily estimate the required amount of funds for the plurality of ATMs as a whole.

Further, ATMs having similar tendencies in money are collected and grouped to create a group characteristic, and this group characteristic is combined with a knowledge characteristic based on a day of the week and a date, so that each of the ATMs has a required characteristic. By estimating the amount of funds, more appropriate funds can be predicted at appropriate times. As a result, the efficiency of fund management can be improved, and the financial institution can obtain the effect of reducing funds and also reduce the number of workers.

[0146]

As described in detail above, according to the present invention, it is possible to arbitrarily change the expected range of the required amount of money using a graph, for example, and to supplement or collect cash on the day before a holiday or at a branch. A fund prediction method and a cash management system that can flexibly respond to timing can be provided.

Further, according to the present invention, it is possible to provide a fund prediction method and a cash management system which can easily predict the required amount of funds of a plurality of automatic transaction apparatuses as a whole.

[Brief description of the drawings]

FIG. 1 is a conceptual diagram of a cash management system in a bank according to an embodiment of the present invention.

FIG. 2 is a block diagram schematically showing a configuration of a part that performs group recycling management using a robot.

FIG. 3 is a perspective view more specifically showing a configuration of a part that performs group recycling management using a robot.

FIG. 4 is a view for explaining an outline of a fund prediction, and shows a place where cash is present in a branch office.

FIG. 5 is a characteristic diagram of the accumulated required fund amount of the automatic transaction device in a branch.

6 is an arrangement diagram of the automatic transaction apparatus corresponding to the characteristic diagram of FIG.

FIG. 7 is a tree diagram of knowledge characteristics created based on days of the week and dates.

FIG. 8 is a diagram showing an example of data used to create the tree diagram of FIG. 7;

FIG. 9 is a characteristic diagram showing a change in accumulated deposits for one month.

FIG. 10 is a characteristic diagram showing a change in the accumulated withdrawal amount for one month.

FIG. 11 is a view for explaining grouping of the automatic transaction apparatus.

FIG. 12 is a diagram for explaining effectiveness by grouping;

FIG. 13 is a functional configuration diagram of a portion for performing fund prediction in a cash management server in a branch.

FIG. 14 is a block diagram showing a configuration of a prediction unit.

FIG. 15 is a flowchart illustrating an outline of a fund prediction process.

FIG. 16 is a flowchart for explaining group characteristic creation / update processing.

FIG. 17 is a diagram showing an example of a created group characteristic.

FIG. 18 is a flowchart for describing knowledge characteristic creation / update processing.

FIG. 19 is a diagram showing an example of a created knowledge characteristic.

FIG. 20 is a flowchart for explaining inference engine creation / update processing.

FIG. 21 is a diagram illustrating an example of a created inference engine.

FIG. 22 is a flowchart illustrating a fund prediction process.

FIG. 23 is a diagram showing an example of various data necessary for fund prediction.

FIG. 24 is a diagram for explaining setting of a margin amount and transition of money amount.

FIG. 25 is a flowchart for explaining a fund prediction process until data accumulates for one month.

FIG. 26 is a diagram for explaining a fund prediction process until data is accumulated for one month.

FIG. 27 is a flowchart illustrating a flow from system introduction to daily operation.

FIG. 28 is a diagram showing a current date and time setting screen.

FIG. 29 is a diagram showing a contact business day setting screen.

FIG. 30 is a diagram showing an ATM operation time zone setting screen.

FIG. 31 is a diagram showing an ATM installation status setting screen.

FIG. 32 is a diagram showing a replenishment / collection instruction unit setting screen.

FIG. 33 is a diagram showing a margin setting screen.

FIG. 34 is a flowchart for explaining the flow during daily operation of the system.

FIG. 35 is a diagram showing a whole banknote graph screen.

FIG. 36 is a view for explaining a method of creating a whole banknote graph in the case of replenishment;

FIG. 37 is a diagram showing a money amount transition graph actually displayed in a whole bill graph in the case of replenishment.

FIG. 38 is a view for explaining a method of creating a whole banknote graph in the case of collection.

FIG. 39 is a diagram showing a money amount transition graph that is actually displayed in a whole banknote graph in the case of collection.

FIG. 40 is a diagram showing another example of a display screen when the funds are exhausted.

FIG. 41 is a diagram showing an example of table data in which money in each denomination is stored for each ATM machine.

FIG. 42 is a view showing a replenishment / collection list screen.

FIG. 43 is a flowchart for explaining the flow of prediction monitoring;

[Explanation of symbols]

A: branch office (store), B: unmanned store, C: calculation center, D: centralized center, 1: in-store LAN, 2: automatic transaction equipment (ATM), 3: robot (carrier) ,
4 Robot controller 13 Cash management server 21 LAN in store 22 Automatic transaction device (A
TM), 42: Host computer, 51: LAN within the center, 54: Fund management server, 71: Information collection unit, 72: Group characteristic creation / update unit, 73: Knowledge characteristic creation / update unit, 74 inference engine creation / update unit, 75 prediction unit, 76 operation / display unit, 81
Fund amount predicting unit, 82: graph creating unit, 83: list creating unit.

Claims (27)

[Claims] 1. An automatic transaction apparatus for performing cash transactions, wherein in the fund prediction method for predicting the amount of funds required for the automatic transaction apparatus, an arbitrary date and time for performing the fund prediction is specified. Based on the cash information of the device, the amount of funds required to operate the automatic transaction device from the current time to the specified date and time is predicted, and replenishment of cash to the automatic transaction device is performed according to the predicted amount of funds. A fund forecasting method comprising calculating a quantity or a collected quantity, and displaying the calculated replenished quantity or a collected quantity. 2. A method for estimating the amount of funds required for an automatic transaction apparatus for performing a cash transaction, the method comprising: The business day of the financial institution where the automatic transaction device is installed, the operating hours of the automatic transaction device, the installation status of the automatic transaction device,
While setting the margin amount for the amount of funds, etc., and specifying an arbitrary date and time for performing the fund forecast, based on the various conditions set and the cash information of the automatic transaction device, from the current time to the specified date and time The amount of funds required for operating the automatic transaction device is predicted, and the amount of replenishment or collection of cash for the automatic transaction device is calculated in accordance with the predicted amount of funds. A fund forecasting method characterized by displaying 3. A method for estimating a necessary amount of money for an automatic transaction apparatus for performing a cash transaction, wherein the automatic transaction is performed by designating an arbitrary date and time for performing a cash prediction. Based on the cash information of the device, the amount of funds required to operate the automatic transaction device from the present time to the specified date and time is predicted, and the amount of cash in the automatic transaction device is changed based on the predicted amount of funds. A graph is created, and based on the created cash amount transition graph, the amount of replenishment or collection of cash for the automatic transaction device is calculated, and the created money amount transition graph is displayed, and the calculated amount is calculated. A method of estimating the amount of cash replenished or collected. 4. Creating the cash amount transition graph includes calculating a required amount of money per day based on the predicted amount of money, and adding the calculated amount of money per day. 4. The method according to claim 3, wherein the cash amount transition graph is created by the following method. 5. The designation of the date and time is performed by displaying a specific mark on the time axis indicated by the displayed money transition graph, and moving the displayed mark along the time axis. 4. The method according to claim 3, wherein
Cash management system as described. 6. A method for estimating a required amount of money for each of a plurality of automatic transaction apparatuses for performing cash transactions, wherein a desired date and time for performing a cash prediction is designated. And, based on the cash information of the automatic transaction device, predicting the amount of funds required to operate each of the plurality of automatic transaction devices from the present time to the specified date and time, based on the predicted amount of funds, A fund forecasting method comprising: creating a list indicating the amount of replenishment or collection of cash for each of the plurality of automatic transaction devices; and displaying the created list of the amount of replenishment or collection. 7. The date and time of the specified prediction range is displayed, and the date and time of the prediction range can be arbitrarily set by changing the displayed date and time. Fund forecasting method. 8. The calculated cash replenishment amount is displayed in a first color, and the calculated cash collection amount is displayed in a second color different from the first color. 7. The fund forecasting method according to claim 6, wherein 9. A method for estimating a required amount of money for each of a plurality of automatic transaction devices for performing a cash transaction, comprising: Manage cash information such as high, specify an arbitrary date and time to make a fund forecast, based on the cash information of each of the plurality of managed automatic transaction devices, from the current time to the specified date and time The amount of funds required to operate each of the plurality of automatic transaction devices is predicted, and the amount of cash replenishment or collection of each of the plurality of automatic transaction devices is calculated in accordance with the predicted amount of funds. A cash replenishment amount or a collection amount is displayed, and based on the predicted fund amount, cash amount transition data of each of the plurality of automatic transaction devices is created. Monitoring the inventory amount transition data, when said the predicted inventory amount remained with the state of the abnormality in different inventory amount transition funds prediction method characterized by warning to that effect to the attendant. 10. A cash management system for predicting and managing a required amount of funds of each of a plurality of automatic transaction devices for performing a cash transaction, wherein each of the plurality of automatic transaction devices has Management means for managing cash information such as the amount of money in cash; designation means for designating an arbitrary date and time for estimating funds; cash information for each of the plurality of automatic transaction devices managed by the management means A forecasting means for estimating a required amount of money when each of the plurality of automatic transaction devices is operated from the present time to a date and time designated by the designating means; Calculating means for calculating the amount of cash replenishment or collection of each of the plurality of automatic transaction devices; and displaying the amount of cash replenishment or collection calculated by the calculation means. Cash management system characterized by comprising a display unit, the that. 11. A cash management system for predicting and managing a required amount of funds for each of a plurality of automatic transaction devices for performing a cash transaction, comprising: Management means for managing cash information such as cash balance, current date and time as various conditions for performing fund prediction, business days of the financial institution where the automatic transaction device is installed, operating hours of the automatic transaction device, Installation status of the automatic transaction device,
Condition setting means for setting a margin amount or the like with respect to the fund amount; specifying means for specifying an arbitrary date and time for performing fund forecasting; various conditions set by the condition setting means; Based on the cash information of each of the plurality of automatic transaction devices, based on the cash information, a prediction unit for predicting the amount of funds required to operate each of the plurality of automatic transaction devices from the current time until the date and time designated by the designation unit Calculating means for calculating the replenishment or collection amount of cash for each of the plurality of automatic transaction devices according to the fund amount predicted by the prediction means; and the replenishment or collection amount of cash calculated by the calculation means A cash management system, comprising: display means for displaying the following. 12. The cash management system according to claim 11, wherein said condition setting means further includes a unit setting means for setting a unit of a replenishment / collection amount of cash as various conditions. 13. The predicting means calculates a required amount of funds of each of the plurality of automatic transaction devices based on cash information of each of the plurality of automatic transaction devices based on various conditions set by the condition setting device. Based on first forecasting means for predicting and various conditions set by the condition setting means, a daily reference fund amount is calculated based on cash information of each of the plurality of automatic transaction apparatuses, and the calculated reference value is calculated. Second prediction means for predicting the required amount of money of each of the plurality of automatic transaction devices based on the amount of money, period determination means for determining a period from the introduction of the cash management system, and determination results of the period determination means; If only a predetermined period has elapsed since the introduction of the cash management system, the second predicting means is used to predict the amount of funds, and after the introduction of the cash management system a predetermined period or more has elapsed. 12. The cash management system according to claim 11, further comprising: a prediction control unit configured to predict the amount of funds using the first prediction unit when the time has passed. 14. When a condition change, such as a change in the installation status of an automatic transaction apparatus, is made, the condition is set again by the condition setting means, and then the amount of funds is set by the second and first prediction means. 14. The cash management system according to claim 13, wherein prediction is performed sequentially. 15. A cash management system for predicting and managing a required amount of funds for each of a plurality of automatic transaction devices for performing a cash transaction, comprising: Management means for managing cash information such as the amount of money in cash; designation means for designating an arbitrary date and time for estimating funds; cash information for each of the plurality of automatic transaction devices managed by the management means A forecasting means for estimating a required amount of money when each of the plurality of automatic transaction devices is operated from the present time to the date and time designated by the designating means; and Processing means for creating a cash amount transition graph of the automatic transaction device; and Calculating means for calculating the replenishment amount or collection amount of cash, and displaying the cash amount transition graph created by the processing means, and calculating the replenishment amount or collection amount of cash calculated by the calculation means. A cash management system, comprising: display means for displaying. 16. The processing means according to claim 1, further comprising: creating a cash amount transition graph for the plurality of automatic transaction devices as a whole;
Alternatively, the automatic transaction device includes a selection unit for selecting and instructing creation of a cash amount transition graph of each of the plurality of automatic transaction devices. The display unit includes an automatic transaction for creating a cash amount transition graph by the selection unit. The cash management system according to claim 15, further comprising a selection designation display unit for designating a device. 17. The first processing unit according to claim 1, wherein the processing unit creates a cash amount transition graph for the entire plurality of automatic transaction devices.
16. The cash management system according to claim 15, further comprising: a processing unit configured to generate a cash amount transition graph of each of the plurality of automatic transaction devices. 18. The display means, wherein when the first processing means displays a cash amount transition graph for a plurality of automatic transaction apparatuses as a whole, if there is an automatic transaction apparatus which has run out of funds. 18. The cash management system according to claim 17, further comprising a message display unit for simultaneously displaying the information by a message. 19. The processing means obtains a required amount of money for each day on the basis of the amount of money predicted by the predicting means, and adds the obtained amount of money for each day to obtain a cash amount. 17. The cash management system according to claim 15, wherein a money amount transition graph is created. 20. The designation means displays a specific mark on the time axis indicated by the money amount transition graph displayed on the display means, and moves the displayed mark along the time axis. 16. The cash management system according to claim 15, further comprising date and time specifying means for specifying the date and time of the cash management. 21. The cash according to claim 20, wherein said date and time designation means has a movement control means for changing a display range by moving a displayed scroll bar along its time axis. Management system. 22. A cash management system for predicting and managing a required amount of funds of each of the plurality of automatic transaction devices for a plurality of automatic transaction devices for performing cash transactions, comprising: Management means for managing cash information such as the amount of money in cash; designation means for designating an arbitrary date and time for estimating funds; cash information for each of the plurality of automatic transaction devices managed by the management means A forecasting means for estimating a required amount of money when each of the plurality of automatic transaction devices is operated from the present time to a date and time designated by the designating means; Calculating means for calculating the amount of cash replenishment or collection of each of the automatic transaction devices; and replenishment of cash for each of the automatic transaction devices calculated by the calculation means. Alternatively, there is provided: a creating unit that creates a list indicating the collection amount; and a display unit that displays a list indicating the replenishment amount or the collection amount of cash of each of the automatic transaction devices created by the creating unit. A cash management system characterized by the following. 23. The display unit, further comprising: a prediction range display unit for displaying a date and time of the prediction range designated by the designation unit, wherein the designation unit includes a date and time displayed on the prediction range display unit. 23. The cash management system according to claim 22, further comprising a date and time setting unit that enables the date and time of the prediction range to be arbitrarily set by changing. 24. The display device, comprising: a cash replenishment amount display processor for displaying the replenishment amount calculated by the calculator in a first color; and a cash collection amount calculated by the calculator. 23. The cash management system according to claim 22, further comprising cash collection amount display processing means for displaying a second color different from the first color. 25. A cash management system for predicting and managing a required amount of money for each of a plurality of automatic transaction devices for performing a cash transaction, comprising: Management means for managing cash information such as the amount of money in cash; designation means for designating an arbitrary date and time for estimating funds; cash information for each of the plurality of automatic transaction devices managed by the management means A forecasting means for estimating a required amount of money when each of the plurality of automatic transaction devices is operated from the present time to a date and time designated by the designating means; Calculating means for calculating the amount of cash replenishment or collection of each of the plurality of automatic transaction devices; and displaying the amount of cash replenishment or collection calculated by the calculation means. Display means for displaying, based on the amount of funds predicted by the predicting means, a means for generating cash transition data for each of the plurality of automatic transaction devices, and a means for generating cash transition data. Monitoring the money status transition data created in the above, when the money status transition is abnormally different from the money status transition predicted by the prediction means, a prediction monitoring means for alerting a staff to that effect; A cash management system comprising: 26. A predicting means, comprising: setting means for setting a reserve amount as a margin in advance; and a prediction process for predicting a necessary amount of funds of the automatic transaction apparatus using the reserve amount set by the setting means. The prediction monitoring means has warning means for warning a clerk when the cash amount of the automatic transaction apparatus is smaller than the reserve amount set by the setting means. 26. The cash management system according to claim 25, wherein: 27. A cash management system for managing an operation state including a cash amount of a plurality of automatic transaction devices installed in a store of a financial institution, wherein the cash management system is installed in the store. Has a group management function with a carrier that stores cash to be replenished or collected for the plurality of automatic transaction devices and moves between the plurality of automatic transaction devices, and in a store equipped with the carrier, A cash management system for managing cash of a plurality of automatic transaction devices using a group management function, comprising: a management unit for managing cash information such as a cash amount of each of the plurality of automatic transaction devices; and Specifying means for specifying an arbitrary date and time; setting means for setting a first reserve amount as a margin in advance to perform funding prediction; and Based on the cash information of each mobile transaction device and the reserve amount set by the setting means, the amount of funds required to operate each of the plurality of automatic transaction devices from the current time until the date and time specified by the specification means And a fund forecasting means for calculating a replenishment amount or a collection amount of cash of each of the plurality of automatic transaction devices according to the forecasted money amount. The fund forecasting means comprises: When calculating the amount of replenishment or collection of cash for each transaction device, first calculation means for calculating the amount of replenishment or collection of cash using the first reserve amount set by the setting means; When calculating a replenishment amount or a collection amount of cash as a whole of the plurality of automatic transaction devices in a store where the carrier is not installed, the amount of cash replenishment or the collection amount is determined based on the first reserve amount set by the setting means. A second reserve amount obtained by adding the number of the plurality of automatic transaction devices installed, and using the second reserve amount to calculate a replenishment amount or a collection amount of cash; When calculating the amount of replenishment or collection of cash as a whole of the plurality of automatic transaction devices at the store where the carrier is installed, a third reserve amount smaller than the second reserve amount is obtained, and And a third calculating means for calculating a replenishing amount or a collecting amount of the cash using the third reserve amount.