A Method, System and Storage Medium for Cash Allocation of an ATM Device

Abstract

The present invention provides a method, a system and a storage medium for dispensing banknotes for an ATM device, which are characterized in that according to the relationship among the balance Y of the cash box, the number of working days X and the daily average change amount K of the cash box balance: Y = K * X + B, where B is the original balance of the cash box; according to the predicted change in the cash box balance of each ATM device over a period of time in the future and the cash clearing date of each ATM device, the frequency of cash replenishment required within each cash clearing cycle of each ATM device and the amount of cash replenishment each time are automatically generated, and a cash dispensing plan is automatically generated. It can achieve relatively accurate automatic prediction of the time and quantity of cash dispensing required for each ATM, reduce a large amount of repetitive manual work, reduce the equipment operation and maintenance costs, and also improve the equipment operation efficiency.

Description

Technical Field

[0001] The present invention relates to the field of financial self-service equipment applications, and more specifically to a method, system, and storage medium for dispensing banknotes in an ATM device. Background Art

[0002] ATM machines are set up to facilitate users to deposit and withdraw banknotes, and their physical layout is as extensive as possible, which means they are very scattered. Therefore, the maintenance workload consumes a large amount of manpower and physical resources. In particular, it is necessary to ensure that there is sufficient banknotes stored in the ATM machines to provide services. Therefore, it is necessary to dispense banknotes to each ATM machine in a timely manner. Since the usage conditions of ATMs at different outlets are different, the dispensing frequencies and quantities consumed and required are different. If non-discriminatory dispensing maintenance is carried out, the efficiency is obviously low. Therefore, it is necessary to formulate an appropriate dispensing plan according to the actual situation to improve the efficiency. The traditional method is to manually count the general rules of ATM devices and deduce the dispensing plan for the next week. The problem with this is that the rules counted manually may not be accurate, there are differences in the rules counted and evaluated by different people, and it takes time. Summary of the Invention

[0003] The technical problem to be solved by the present invention is how to improve the prediction accuracy of the banknote dispensing requirements of each ATM device and improve the accuracy and rationality of formulating the dispensing plan.

[0004] To solve the above problems, the present invention provides a method for dispensing banknotes in an ATM device, which is characterized in that it is assumed according to experience that the change in the balance Y of the cash box is basically linearly related to the number of working days X and the average daily change amount K of the cash box balance of each ATM device. Specifically: Y = K * X + B, where B is the original cash box balance; the system database separately maintains the operation data of each ATM device and each cash box; the average daily change amount K of the cash box balance is different at different positions, and the average daily change amount K of the cash box balance is obtained by statistically analyzing the historical actual operation data of each ATM device; according to the predicted change in the cash box balance of each ATM device in the future for a period of time and the cash clearing date of each ATM device, the frequency of adding banknotes and the quantity of each banknote addition within each cash clearing cycle of each ATM device are automatically generated, and a dispensing plan is automatically generated. Specifically:

[0005] Step 1: Determine the next cash clearing date of each ATM device respectively: the previous banknote addition date + N natural days. Determine whether the last day is a non-working day. If so, the next cash clearing deadline date = the first 2 working days before the N-day cash clearing date. If not, the next cash clearing deadline date = the first 3 working days before the 10-day cash clearing date;

[0006] Step 2: Determine the banknote addition and dispensing dates respectively:

[0007] Step 1: Calculate the remaining available days D of the surplus cash in the predicted self-service banking points. There are i ATM devices in each self-service banking point:

[0008] Let D1 be the remaining available days of the cash box of a single ATM device. D1 = balance of the cash box / K or (limit of the cash box - balance of the cash box) / K;

[0009] Let ∑Ki be the total daily change in the balance of the cash boxes of each device in the self-service banking point, and let ∑Mi be the total predicted cash replenishment amount of each device in the self-service banking point;

[0010] Let D2 be the remaining available days of the cash boxes in the self-service bank. D2 = 0 or 10, and the value is determined by ∑Ki and ∑Mi in the self-service banking point according to the following method:

[0011] If ∑Ki > 0, then the deposits in the self-service banking point are greater than the withdrawals, and the balance tends to increase. If ∑Mi > 400,000 × i, then D2 = 0; otherwise, D2 = 10;

[0012] If ∑Ki < 0, then the withdrawals in the self-service banking point are greater than the deposits, and the balance tends to decrease. If ∑Mi < 30,000 × i, then D2 = 0; otherwise, D2 = 10;

[0013] Let D3 be the anti-duplicate cash replenishment control variable. If the current date - the previous cash replenishment date ≤ 3 days, then D3 = 10; if the current date - the previous cash replenishment date > 3 days, then D3 = 0;

[0014] Finally, obtain the predicted available days D = D1 + D2 + D3;

[0015] Predicted cash replenishment date = current date + predicted available days;

[0016] Determine the actual planned cash replenishment date, and take the smaller value between the predicted cash replenishment date and the next cash clearing deadline date;

[0017] Automatically generate a cash replenishment plan form.

[0018] For the cash replenishment method of the ATM device described above, determine the amount of each cash replenishment M according to the following method. Determine the cash replenishment amount according to the different predicted cash replenishment frequencies within each cash clearing cycle. Specifically as follows: For the first category, when the cash replenishment frequency <= 3 and the daily change in the balance of the cash box of the device K is positive, the cash replenishment amount M is equal to 0 yuan; when the daily change in the balance of the cash box of the device K is negative, the cash replenishment amount M is equal to 400,000 yuan;

[0019] For the second category, when the cash replenishment frequency > 3 and the cash replenishment days <= 3, the cash replenishment amount remains the same as the previous time.

[0020] Category 3: When the cash replenishment frequency > 3 and the cash replenishment days > 3, when the daily average change amount K value of the cash box balance is negative, the amount in the cash box tends to decrease. Let M = -K × 10 and round up to the nearest multiple of 50,000. The maximum value of M is 400,000. When the K value is positive and less than or equal to 20,000, let M = 50,000. When the K value is positive and greater than 20,000, let M = 0.

[0021] The cash allocation method of the ATM device described above is characterized in that for the daily average change amount K value of the cash box balance, it will be estimated according to the actual cash usage rules of each machine in history every year, and different daily average change amount K values of the cash box balance will be used for estimation in different cash clearing cycles for the same ATM device.

[0022] The cash allocation method of the ATM device described above is characterized in that the maximum cash allocation amount and the maximum number of self-service bank branches with cash allocation per day are set for each working day; when the maximum cash allocation amount is exceeded or the maximum number of self-service bank branches with cash allocation per day is exceeded, the available balances of each self-service bank branch are automatically sorted from high to low according to the prediction, and some self-service bank branches with high available balances are arranged for cash allocation on the next working day.

[0023] An ATM device cash allocation system, characterized in that it is a cash allocation system implemented by using the above-mentioned ATM device cash allocation method.

[0024] A computer-readable storage medium, characterized in that the readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor is caused to execute the above-mentioned ATM device cash allocation method.

[0025] Implementing the present invention has the following beneficial effects: It can realize relatively accurate automatic prediction of the time and quantity of cash allocation required for each ATM machine, reduce a large amount of repetitive manual work, reduce the equipment operation and maintenance cost, and also improve the equipment operation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a flow chart of the cash allocation method of the ATM device. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0028] In view of the problems existing in the above-mentioned prior art, the present invention proposes a method for dispensing banknotes in an ATM device, which searches for the hidden rules from a large amount of balance data of the cash boxes of self-service devices and the banknote replenishment regulations of self-service devices through a linear regression algorithm, so as to predict the banknote replenishment plan of the ATM self-service device. The intelligent generation of the banknote replenishment plan is carried out by using a linear regression model and SQL technology. Taking a 10-day cleaning as an example as a banknote replenishment cycle according to the balance of the cash box and the cleaning rules, the linear regression of the change rule between the balance of the cash box and time is carried out according to the change rule of the balance of the cash box after each banknote replenishment, and the next banknote replenishment date and amount are predicted based on the slope (K) of the straight line and the value range of the standard deviation of the error.

[0029] As a systematic management, it is actually necessary to be equipped with the management of other complete basic databases. There are different ATM machines and different cash boxes in different network points. Therefore, it is necessary to carry out separate maintenance for each ATM device, including the basic data of the device, the maintenance data of the device, the historical operation data of the device, etc. The following examples illustrate some conventional implementation methods of basic database management:

[0030] (1) Management of the list of received devices: The content of the "List of Received Devices" includes: receipt date, network point name, network point institution number, address, machine number, device type, device brand, model, deposit and withdrawal function, operation status, change date. The management functions include: import, new addition, modification, and query. Import function: It refers to supporting the upload of data files. New addition function: It refers to supporting the entry of a single device into the warehouse. Modification function: It refers to supporting the modification of the device's affiliated institution and operation status. The operation status includes running, withdrawal of the point, withdrawal of the machine, and scrapping. The modification records can be traced and queried. Query function: It refers to supporting the screening query of the list of received devices according to the form elements, including historical modification records.

[0031] (2) Import of the "Relationship between Model and Cash Box Capacity": It includes model, number of cash boxes, cash box capacity, and device capacity. The relationship is device capacity = number of cash boxes * cash box capacity, which is used to judge the maximum limit of banknote replenishment and dispensing for each device.

[0032] (3) Import of the "Work and Rest Time Arrangement": It includes date, holiday name, work and rest arrangement, compensatory leave arrangement, and banknote replenishment and dispensing coefficient. The work and rest arrangement includes two states: work and rest. The compensatory leave arrangement has two states: compensatory leave and no compensatory leave. If a legal holiday is for compensatory leave, it is regarded as a working day. The banknote replenishment and dispensing coefficient is based on 1. For example, before the Spring Festival and National Day, the cash withdrawal demand increases, and the coefficient can be adjusted upwards to increase the cash supply. During the festival, the deposit demand increases, and the coefficient can be adjusted downwards to reduce the cash supply.

[0033] (4) Management of the "Work Order Information": Import the downloaded "Work Order Information" and the registration form of the "Temporary Work Order" after the order is executed into the database.

[0034] (5) "Cash Box Balance" Management: Import the downloaded "Cash Box Balance" into the database.

[0035] (6) "Communication Status" Management: Import the obtained "Communication Status" into the database.

[0036] The core of the present invention lies in automatically generating a cash addition and replenishment plan based on basic data and a linear model. The following is an implementation description based on a 10-day cash clearing rule:

[0037] Cash Addition and Replenishment Plan Generation:

[0038] (1) Cash addition and replenishment model. According to the cash box balance and the 10-day cash clearing rule, with one cash addition and replenishment as a cycle, based on the change law of the cash box balance after each cash addition and replenishment, perform a linear regression on the change law of the cash box balance and time to obtain the daily average change amount K of the cash box balance for each device.

[0039] The cash addition and replenishment model adopts a linear regression model and is implemented using numpy in python. Assuming that the change of each device is linear, select the straight line equation Y = K*X + B that is closest. Y is the cash box balance of the device, X is the number of days, and fit the daily average change amount K (i.e., the slope) and its intercept B (B is the original cash box balance) of the cash box balance for each device, and perform relevant calculations based on K.

[0040] (2) Next cash clearing deadline date.

[0041] First step: Calculate the 10-day cash clearing date = the previous cash addition date + 10 natural days.

[0042] Second step: Determine whether this date is a working day.

[0043] Third step: Make adjustments according to the judgment result. If it is, the next cash clearing deadline date = the two working days before the 10-day cash clearing date; if not, the next cash clearing deadline date = the three working days before the 10-day cash clearing date.

[0044] (3) Determination of cash addition and replenishment amount.

[0045] Let M be the predicted cash addition amount of the device.

[0046] First category: When the cash addition frequency <= 3 and the daily average change amount K of the cash box balance of the device is positive, the cash addition amount M is equal to 0 yuan; when the daily average change amount K of the cash box balance of the device is negative, the cash addition amount M is equal to 400,000 yuan.

[0047] Second category: When the cash addition frequency > 3 and the cash addition days <= 3, the cash addition amount remains the same as the previous time.

[0048] Category 3: When the cash replenishment frequency > 3 and the cash replenishment days > 3, when the daily average change amount K of the cash box balance is negative, the amount in the cash box tends to decrease. Let M = -K × 10 and round up to the nearest multiple of 50,000, with the maximum value of M being 400,000. When K is positive and less than or equal to 20,000, let M = 50,000. When K is positive and greater than 20,000, let M = 0.

[0049] (4) Determination of the cash replenishment and dispensing date

[0050] Step 1: Calculate the predicted available days D. Let D1 be the remaining available days of the cash box for a single device. D1 = cash box balance / K or (cash box limit - cash box balance) / K;

[0051] Let ∑Ki be the sum of the daily average change amounts of the cash box balances of each device in the self-service bank. Let ∑M i be the sum of the predicted cash replenishment amounts of each device in the self-service bank. Let D2 be the remaining available days of the self-service bank cash box. D2 = 0 or 10, and the value is determined by ∑Ki and ∑M i of the self-service bank.

[0052] If ∑Ki > 0, then the deposits in the self-service bank are greater than the withdrawals, and the balance tends to increase. If ∑M i > 400,000 × number of devices, then D2 = 0; otherwise, D2 = 10.

[0053] If ∑Ki < 0, then the withdrawals in the self-service bank are greater than the deposits, and the balance tends to decrease. If ∑M i < 30,000 × number of devices, then D2 = 0; otherwise, D2 = 10.

[0054] Let D3 be the anti-duplicate cash replenishment control variable. If the current date - the previous cash replenishment date ≤ 3 days, then D3 = 10. If the current date - the previous cash replenishment date > 3 days, then D3 = 0.

[0055] Finally, the predicted available days D = D1 + D2 + D3.

[0056] Step 2: Calculate the predicted cash replenishment and dispensing date. Predicted cash replenishment and dispensing date = current date + predicted available days.

[0057] Step 3: Calculate the actual cash replenishment and dispensing date. Actual cash replenishment and dispensing date = the smaller value between the predicted cash replenishment and dispensing date and the next cash clearing deadline date.

[0058] (5) Finally, automatically generate the "Cash Replenishment and Dispensing Plan" table.

[0059] Comprehensively apply the "List of Equipment to be Withdrawn", "Table of Relationship between Model and Cash Box Capacity", "Historical Cash Loading Plan", "Balance Information Table", and "Work and Rest Schedule" to generate the "Cash Loading and Allocation Plan". Generate the cash loading and allocation plans for each working day within 10 natural days before 11:30 am every working day, and distribute the equipment as evenly as possible across each day; the total balance of the equipment is controlled within 140 million yuan, except before holidays. When a certain piece of equipment at a self-service bank branch meets the cash loading condition, include all the equipment at that branch in the cash loading plan for the same day according to the cash loading rules. A daily ATM self-service equipment table can be generated, which includes the fields: machine number, equipment address, the most recent cash loading time, expiration date, predicted cash loading amount, affiliated institution, slope K. It is possible to determine whether the day 8 days after the most recent cash loading time of each piece of equipment is a working day based on the work and rest status table. If it is a working day, that date is the expiration date; if it is not a working day, extract one or two days. An ATM equipment cash box balance cycle information table required for model establishment can be generated. This table includes the fields: machine number, cash box balance, date, date serial number; after sorting the dates in ascending order and numbering them, then call the linear regression model to calculate the slope K. It is possible to calculate the available days and connect to the work and rest status table to obtain the corresponding working days. Select the machines that must be cash loaded the next day and add other equipment of the same institution to the plan for the next day. To comprehensively achieve the balance of daily cash loading machines, calculate the number of equipment to be cash loaded in the plan generated on the same day. If the number of equipment is less than the total number of equipment divided by 8 days, sort the equipment in ascending order of the predicted cash loading date, select the first piece of equipment, and then add other equipment of the institution to which this piece of equipment belongs to the plan. If the number of equipment is still small, repeat the above process

[0060] In summary, the above embodiments are only illustrative of the principles and effects of the present invention, and are not used to limit the present invention. Any person familiar with this technology can modify or change the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or changes made by those with ordinary knowledge in the technical field without departing from the spirit and technical ideas disclosed by the present invention should still be covered by the claims of the present invention.

Claims

1. A method for dispensing banknotes in an ATM device, characterized in that Based on empirical assumptions, the change in the balance Y of the cash cassette is linearly related to the number of working days X and the average daily change amount K of the cash cassette balance of each ATM device. Specifically: Y = K * X + B, where B is the original cash cassette balance; the system database separately maintains the operation data of each ATM device and each cash cassette; the average daily change amount K of the cash cassette balance is different at different locations, and the average daily change amount K of the cash cassette balance is obtained by statistically analyzing the historical actual operation data of each ATM device; according to the predicted change in the cash cassette balance of each ATM device in the future period and the cash clearing date of each ATM device, the frequency of cash replenishment required for each ATM device within each cash clearing cycle and the amount of each cash replenishment are automatically generated, and a cash allocation plan is automatically generated. Specifically: Step 1: Determine the next cash clearing date for each ATM device respectively: the previous cash replenishment date + N natural days, and determine whether the last day is a non-working day. If it is, the next cash clearing deadline date = the first 2 working days before the N-day cash clearing date; if not, the next cash clearing deadline date = the first 3 working days before the 10-day cash clearing date. Step 2: Determine the cash replenishment and allocation date respectively: The first step: Calculate the predicted available days D of the remaining cash in the self-service banking point. There are i ATM devices in each self-service banking point: Let D1 be the remaining available days of the cash cassette of a single ATM device, D1 = cash cassette balance / K or (cash cassette limit - cash cassette balance) / K; Let ∑Ki be the total of the average daily change amounts of the cash cassette balances of each device in the self-service banking point, and let ∑Mi be the total of the predicted cash replenishment amounts of each device in the self-service banking point; Let D2 be the remaining available days of the self-service banking cash cassette, D2 = 0 or 10, and the value is determined by ∑Ki and ∑Mi in the self-service banking point according to the following method: If ∑Ki > 0, then the deposits in the self-service banking point are greater than the withdrawals, and the balance tends to increase. If ∑Mi > 400,000 × i, then D2 = 0; otherwise, D2 = 10; If ∑Ki < 0, then the withdrawals in the self-service banking point are greater than the deposits, and the balance tends to decrease. If ∑Mi < 30,000 × i, then D2 = 0; otherwise, D2 = 10; Let D3 be an anti-duplicate cash replenishment control variable. If the current date - the previous cash replenishment date ≤ 3 days, then D3 = 10; if the current date - the previous cash replenishment date > 3 days, then D3 = 0; Finally, obtain the predicted available days D = D1 + D2 + D3; The predicted cash replenishment and allocation date = the current date + the predicted available days; Determine the actual planned cash replenishment and allocation date, and take the smaller value of the predicted cash replenishment and allocation date and the next cash clearing deadline date; Automatically generate a cash allocation plan form.

2. The method for dispensing banknotes of an ATM device according to claim 1, wherein Determine the amount M of each cash replenishment and allocation according to the following method. Determine the cash replenishment amount according to different predicted cash replenishment frequencies within each cash clearing cycle. Specifically as follows: The first category, when the cash replenishment frequency <= 3, and the average daily change amount K of the cash cassette balance of the device is positive, the cash replenishment amount M is equal to 0 yuan; when the average daily change amount K of the cash cassette balance of the device is negative, the cash replenishment amount M is equal to 400,000 yuan. The second category, when the cash replenishment frequency > 3 and the cash replenishment days <= 3, the cash replenishment amount is the same as the previous time; Category 3: When the cash replenishment frequency > 3 and the cash replenishment days > 3, when the daily average change amount K value of the cash box balance is negative, the amount in the cash box tends to decrease. Let M = -K × 10 and round up to the nearest multiple of 50,000, with the maximum value of M being 400,000; when the K value is positive and less than or equal to 20,000, let M = 50,000, and when the K value is positive and greater than 20,000, let M = 0.

3. The method for dispensing banknotes of the ATM device according to claim 2, wherein For the daily average change amount K value of the cash box balance, it will be estimated with different daily average change amount K values of the cash box balance in different cash clearing cycles for the same ATM device according to the actual cash usage pattern of each device every year.

4. The method for dispensing banknotes of an ATM device according to claim 3, wherein Set the maximum cash allocation amount and the maximum number of self-service bank outlets with cash allocation per day for each working day; when the maximum cash allocation amount is exceeded or the maximum number of self-service bank outlets with cash allocation per day is exceeded, automatically sort the available balances of each self-service bank outlet predicted from high to low, and arrange some self-service bank outlets with high available balances for cash allocation on the next working day.

5. A banknote dispensing system for an ATM device, characterized in that, A cash allocation system implemented by using the cash allocation method of the ATM device according to any one of claims 1 to 4.

6. A computer-readable storage medium, characterized in that, A readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor is caused to execute the cash allocation method of the ATM device according to any one of claims 1 to 4.

Images