JP7513037B2 - Change management system, change management method and program - Google Patents

Description

The present invention relates to a change management system, a change management method, and a computer-readable medium.

In the retail and food service industries, the change (denomination) reserves to be kept in the change dispensers of cash registers (POS registers) installed in stores are prepared by the sales headquarters or store operators (store managers, etc.) at their own discretion, based on past trends, experience, intuition, etc. However, there are frequent cases of excess or shortage of change during store operations. As a countermeasure, employees always check the remaining number of coins of each denomination and the contents of the change dispenser displayed on the cash register's change dispenser before they run out of change, and store managers use this prediction to make individual decisions about exchanging money before a shortage occurs.

After deciding to exchange, the store manager or employee takes out money such as 10,000 yen bills from the cash register and exchanges them at a bank, etc., to ensure that there is no shortage of change. However, even if a shortage is predicted and exchange is made, the exchanged denomination is not always used, and it may be unnecessary (unused). On the other hand, if there is a possibility of an excess (cash overflowing from the cash register), the relevant denomination is withdrawn from the cash register's change machine by visually checking the situation inside the change machine or by an alarm from the cash register, and the cash is stored in the safe until the sales amount is collected. The above exchange and storage work requires the manpower of employees and incurs exchange fees. It is also an unnecessary task outside of the normal work. In addition, the surplus money is left in the store, which increases the security risk. Furthermore, there is also the problem that stores that borrow reserve funds have to pay extra interest due to the existence of excess cash, and stores that operate with their own funds cannot use funds that should be invested.

For this reason, systems that predict the amount of change to be replenished in cash register change dispensers are being considered. Patent Document 1 discloses a system that accumulates currency and banknote data in a database, including the number of banknotes or coins received and withdrawn for each transaction and each change denomination at a store, as well as the final change inventory for each day, and then performs a predictive calculation of the appropriate amount of change to be replenished based on the accumulated data.

JP 2013-020582 A

The system described in Patent Document 1 basically only predicts the amount of change to be replenished. However, even if the amount of change to be replenished can be predicted, it does not necessarily mean that the change fund will be distributed to the store when change replenishment is required. Also, there may be cases where the change fund is distributed to the store when change replenishment is not required. For these reasons, it is desirable to predict the appropriate cycle for distributing the change fund.

In consideration of the above-mentioned problems, the purpose of the present disclosure is to provide a change management system that can appropriately calculate the number of coins to be distributed of each denomination and the distribution cycle.

The change management system according to the first aspect of the present invention comprises a POS system, a prediction analysis unit which uses data relating to transactions in the POS system and data for each denomination in a change machine to predict an increase or decrease in change for each denomination for each target store during a target period, and a distribution calculation unit which uses the prediction results from the prediction analysis unit to calculate the number of changes to be distributed for each denomination and the distribution cycle.

The change management method according to a second aspect of the present invention comprises a first step of predicting an increase or decrease in change of each denomination for each target store during a target period using data related to transactions in a POS system and data for each denomination in a change dispenser, and a step of calculating the amount of change to be distributed of each denomination using the prediction results from the first step.

A non-transitory computer-readable medium according to a third aspect of the present invention stores a program that causes a computer to execute a first step of predicting the increase or decrease in change of each denomination for each target store during a target period using data related to transactions in a POS system and data for each denomination in a change dispenser, and a step of calculating the amount of change to be distributed for each denomination using the prediction results from the first step.

According to the present invention, the number of payments for each denomination and the cycle for payments can be appropriately calculated.

1 is a block diagram showing a configuration of a change management system according to a first embodiment; FIG. 11 is a block diagram showing the configuration of a change management system according to a second embodiment. 13 is a flowchart showing a process flow of a change management system according to the second embodiment. 13 is a flowchart showing a process flow of a change management system according to the second embodiment. FIG. 2 is a schematic diagram showing an example of a schedule table.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings. For clarity of explanation, the following description and drawings have been omitted and simplified as appropriate. In each drawing, the same elements are given the same reference numerals, and duplicate explanations have been omitted as necessary.

[First embodiment]
The first embodiment will be described below.
Fig. 1 is a block diagram showing the configuration of a change management system 200 according to the first embodiment. As shown in Fig. 1, the change management system 200 includes a POS system 201, a prediction analysis unit 204, and a distribution calculation unit 212. The prediction analysis unit 204 predicts an increase or decrease in change of each denomination for each target store during a target period, using data related to transactions in the POS system 201 and data of each denomination in the change machine. The distribution calculation unit 212 calculates the number of changes to be distributed for each denomination and the distribution cycle, using the prediction results by the prediction analysis unit 204.

By doing this, the number of payments for each denomination and the payment cycle can be appropriately calculated.

[Embodiment 2]
The second embodiment will now be described.
First, a configuration example of a change management system 100 according to the second embodiment will be described. The change management system 100 is intended to improve the efficiency of store operations related to change. Fig. 2 is a block diagram showing the configuration of the change management system 100 according to the second embodiment. As shown in Fig. 2, the change management system 100 has a POS system 1, a prediction analysis unit 4, a prediction result output unit 7, a distribution unit 10, a distribution calculation unit 12, and an output unit 20.

POS system 1 is a system that manages and analyzes store sales, but here we will mainly explain the parts related to distributing change funds. POS system 1 includes a data extraction unit 2 and a setting unit 3. Data extraction unit 2 accumulates data such as sales transacted in POS system 1 (POS-related data) and various data at the time of transactions of each denomination in the change dispenser (change-related data).

Here, the POS-related data includes the purchased product, purchase date and time, purchase amount, purchase location, number of customers, etc. The change-related data includes the transaction date and time, the number of denominations paid at the time of transaction, and the number of denominations of change received. If a point card is used at the time of transaction, the change-related data is weighted information on deposits and withdrawals of each change categorized from persona data for each store created based on customer information on the point card (gender, age, number of purchases, purchased products, etc.). The POS system 1 is linked to an external system 900. The external system 900 is, for example, a point card information management system. The POS system 1 acquires customer attribute data from the external system 900 and stores it in the data extraction unit 2. If the external system 900 is a point card management system, the data extraction unit 2 stores customer attribute data linked to the point card used in the POS system 1.

The setting unit 3 dynamically calculates the first and second thresholds from information about the storage capacity of the change machine for each target store and the prediction results of the prediction analysis unit 4 described below. Here, the first threshold is the upper limit of the change amount in the change machine for each target store (the amount of change that may overflow), and the second threshold is the lower limit of the change amount (the amount of change that may not be enough). Information about the storage capacity of the change machine for each target store is received from the distribution unit 10 described below. The POS system 1 is configured to issue an alarm when the amount of change exceeds the first threshold or falls below the second threshold.

The prediction analysis unit 4 uses data related to transactions in the POS system 1 and data for each denomination in the change machine to predict increases or decreases in change for each denomination at each target store during the target period, and includes a data collection/processing unit 5 and a prediction unit 6. The data collection/processing unit 5 periodically collects POS-related data and change-related data from the POS system 1 and processes it into data for analysis. Here, the data for analysis is data that can be analyzed by the prediction unit 6, including persona data for each store created based on customer information on point cards, for example. The prediction unit 6 uses the processed analyzable data to predict increases or decreases in change for each denomination at each target store during the target period.

When the data collection/processing unit 5 creates persona data from point card information and the prediction unit 6 uses it to perform analysis, the characteristics of people who visit each time period and each day of the week for each target store are highlighted, and accounting trends are understood in detail. For example, suppose that the persona data for weekday lunch at a store in an office district shows that the customer base is mostly office workers in their 30s, that they tend to order large and reasonable lunches, and that they tend not to give small change when paying. Reasonable items here refer to lunch sets that cost around 500 to 600 yen excluding tax, and small change refers to 50 yen, 10 yen, and 1 yen. It is predicted that this store will give out more 50 yen, 10 yen, and 1 yen as change at lunchtime than other stores. In this way, it is thought that using persona data created from point card information will improve the accuracy of predictions of the increase or decrease in change for each denomination for each target store during the target period.

The prediction result output unit 7, which compiles and processes the prediction results to suit the output destination, includes a compilation unit 8 and a processing unit 9. The compilation unit 8 compiles the output prediction result data for each distribution destination. The processing unit 9 processes the prediction result data after compilation into a form suitable for each distribution destination.

The distribution unit 10 distributes the output data to the distribution destinations, and includes a distribution setting function unit 11. The distribution setting function unit 11 sets the distribution destinations and distribution cycle. The distribution unit 10 sends the data to each distribution destination in accordance with the settings in the distribution setting function unit 11.

The dividend calculation unit 12 includes a data input unit 13, a data unit 14, and a dividend plan generation unit 17.

The data import unit 13 imports prediction result data and exchanges various data with the data unit 14. The data unit 14 stores map information 15 for calculating the distribution route (hereinafter referred to as the "distribution route") and shift information 16 (employee information, delivery vehicle information, etc.) for employee shifts and vehicle allocation. The data import unit 13 obtains the map information 15 and shift information 16 from the data unit 14 and sends them to the distribution plan generation unit 17 together with the prediction result data.

The distribution plan generation unit 17 includes a calculation unit 18 and a generation unit 19. The calculation unit 18 calculates the number of distributions of each denomination and the distribution cycle (hereinafter referred to as the "distribution cycle") using the prediction results by the prediction analysis unit 4. The calculation unit 18 may also calculate a distribution route using the prediction results and map information on which the target stores are posted. Furthermore, the calculation unit 18 may calculate employee shifts and vehicle allocation for distribution operations using the prediction results and information on employees and delivery vehicles. The generation unit 19 generates distribution plan data using information on the optimal number of distributions, distribution cycle, distribution route, employee shifts and vehicle allocation in the calculation unit 18, and sends the distribution plan data to the output unit 20.

The output unit 20 outputs the distribution plan data sent from the generation unit 19, and includes an output setting function unit 21. The output setting function unit 21 sets the distribution destination (by distribution company, etc.) and the format for viewing the plan, etc. Based on the settings in the output setting function unit 21, the output unit 20 formats the distribution plan data to suit the output destination smartphone, display, etc., and outputs it.

Next, the process flow of the change management system 100 will be described. In the following explanation, reference will also be made to FIG. 2 as appropriate.

3 and 4 are flowcharts showing the process flow of the change management system 100. As shown in FIG. 3, first, in the POS system 1, the setting unit 3 determines whether or not the setting of the data extraction conditions has been completed (step S1). Here, the extraction conditions include the target stores of the data to be used for analysis, customer information linked to the point card, the designated period, items, the target of the objective variable (variable to be predicted), and the target of the explanatory variable (variable explaining the objective variable). If the setting has been completed (YES), the process proceeds to step S2, and if the setting has not been completed (NO), i.e., if there are other extraction conditions, the process of step S1 is repeated.

Next, the data collection/processing unit 5 of the predictive analysis unit 4 imports the POS-related data and change-related data extracted by the data extraction unit 2 of the POS system 1 (step S2). Next, the POS-related data and change-related data imported by the data collection/processing unit 5 are merged and processed into analysis data (step S3). From the POS-related information, persona data for each store is created, and from the change-related information, the number of transactions per hour, the number of deposits and withdrawals, etc. are counted. In addition, the persona data is generated by machine learning etc. based on the customer's tendency to visit the store (time of day, number of visits, day of visit), age, gender, purchased products, etc.

Following step S3, the prediction section 6 of the prediction analysis section 4 predicts the change increase or decrease from the analysis data (step S4). The prediction of the change increase or decrease is performed using general-purpose analysis tools such as SAP (SAP for Business Analytics) and SPSS (IBM SPSS statistics), and machine learning (regression analysis, etc.) is performed using data processed for analysis (store information, change denominations, withdrawal and deposit information for each denomination, purchased product information, etc.) as learning data to create a prediction model. The prediction model is made up of prediction formulas, etc., and the number of deposits and withdrawals of each denomination, which is the objective variable, is predicted by applying explanatory variables (explanatory variables: number of deposits and withdrawals of each denomination for each time period) generated from past store deposit and withdrawal information, etc. to the prediction formula and analyzing it to predict the number of deposits and withdrawals of each denomination at each store.

Following step S4, the prediction result output unit 7 aggregates the predicted data (prediction result data) for each store (step S5). That is, in step S5, store information, change machine equipment information (including information on change storage capacity), etc. are added to the prediction result data, and the data is merged for each store.

Following step S5, the processing unit 9 of the prediction result output unit 7 calculates a cycle for distributing data (distribution cycle) based on the device information of the change dispenser (step S6). Next, the processing unit 9 processes the prediction result data collected for each store into data for distribution to the distribution calculation unit 12 for each store (step S7).

Following step S5, the processing unit 9 calculates the above-mentioned alarm thresholds (first and second thresholds) based on the device information of the change dispenser (step S8). The processing unit 9 then processes the calculation results of the alarm thresholds into data to be distributed to the POS system 1 (step S9).

After the processing of steps S7 and S9 is completed, the distribution setting function unit 11 of the distribution unit 10 determines whether a distribution destination has been set (step S10). Specifically, the distribution destination of the data to be distributed to the distribution calculation unit 12 and the distribution destination of the data to be used for distribution to the POS system 1 are set for each store and each POS as "on/off" for each store and each POS. If a distribution destination has been set (YES), processing proceeds to step S12; if a distribution destination has not been set (NO), that is, if there are other settings remaining, processing of step S10 is repeated. Next, the distribution unit 10 distributes data to the distribution calculation unit 12 or the POS system 1 based on the contents set in step S10 (step S12).

Following step S12, as shown in Fig. 4, the setting unit 3 of the POS system 1 automatically sets the alarm thresholds for each denomination based on the received threshold data (step S13). Note that the process for setting the alarm thresholds in the POS system 1 ends here.

Furthermore, following step S12, as shown in FIG. 4, in the distribution calculation unit 12, the data import unit 13 imports the aggregated data (step S14). Next, the data import unit 13 imports information such as a road map and the location of the distribution establishment, and information on employees and delivery vehicles (shift information) from the data unit 14 (step S15). Next, the calculation unit 18 of the distribution plan generation unit 17 classifies the target stores by distribution cycle, and calculates the distance to each target store when the location of the distribution establishment is used as the starting point (step S16).

Next, the calculation unit 18 of the distribution plan generation unit 17 generates a distribution store list for each distribution office (step S17). That is, in step S17, for each distribution cycle, a distribution office is assigned a store in charge according to a predetermined distance from the distribution office and the number of stores, in order of proximity to the distribution office, and the stores are listed for each office. Next, the calculation unit 18 calculates an efficient distribution route for each list, starting from the office for each distribution cycle (step S18).

Following step S18, the generation unit 19 of the distribution plan generation unit 17 maps the stores and routes on a map (step S19). Next, the generation unit 19 matches information on employees and delivery vehicles (shift information) (step S20). That is, in step S20, distribution date information is added to the list of distribution amounts by denomination for each store (prediction result data) and distribution route information. Next, the generation unit 19 generates a distribution schedule from the distribution store list and cycle (step S21). That is, in step S21, the distribution date information of step S20 is matched with employee shift information and vehicle allocation information, and a distribution schedule is generated that further includes information on the person in charge and vehicle number for each distribution date. The distribution schedule also includes information on the amount of distribution to be prepared.

Following step S21, the generation unit 19 combines the distribution schedule data from step S21 with the map information on which the route was mapped in step S19 to generate a schedule table (step S22). An example of the schedule table will be described later. Next, the output setting function unit 21 of the output unit 20 determines whether the output setting has been completed (step S23). Here, the output setting sets the output destination, such as PHONE, TAB, display, paper, etc. If the output setting has been completed in step S23 (YES), the process ends, and if the output setting has not been completed in step S23 (NO), the process of step S23 is repeated.

Figure 5 is a schematic diagram showing an example of a schedule. As shown in Figure 5, the distribution date, dispatch number, and dispatcher are listed, and the list lists the identification number of the distribution box and the amount of each denomination for distribution for each store name. Furthermore, the schedule shows a distribution route map.

As described above, the prediction analysis unit 4 uses data related to transactions in the POS system 1 and data for each denomination in the change machine to predict the increase or decrease in change of each denomination for each target store during the target period. The distribution calculation unit 12 then calculates the number of distributing coins of each denomination and the distribution cycle using the prediction results from the prediction analysis unit 4. In this way, it is possible to appropriately determine the number of distributing coins of each denomination and the distribution cycle so that the change reserve can be distributed without excess or deficiency, without having to rely on the past trends or intuition of the sales headquarters or store operators as in the past.

In the above embodiment, the present invention has been described as a hardware configuration, but the present invention is not limited to this. The present invention can also be realized by having a CPU (Central Processing Unit) execute a program to perform each process.

The program for implementing the above-mentioned recognition camera adjustment method can be stored and provided to a computer using various types of non-transitory computer readable media. Non-transitory computer readable media include various types of tangible storage media. Examples of non-transitory computer-readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R/Ws, and semiconductor memories (e.g., mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAMs (Random Access Memory). The program may also be provided to the computer by various types of transient computer-readable media. Examples of transient computer-readable media include electrical signals, optical signals, and electromagnetic waves. The transient computer-readable medium can provide the program to the computer via a wired communication path such as an electric wire or optical fiber, or via a wireless communication path.

The present invention has been described above with reference to the embodiment, but the present invention is not limited to the above. Various modifications that can be understood by a person skilled in the art can be made to the configuration and details of the present invention within the scope of the invention.

1, 201 POS system 2 Data extraction unit 3 Setting unit 4, 204 Prediction analysis unit 5 Data collection/processing unit 6 Prediction unit 7 Prediction result output unit 8 Counting unit 9 Processing unit 10 Distribution unit 11 Distribution setting function unit 12, 212 Distribution calculation unit 13 Data import unit 14 Data unit 15 Map information 16 Shift information 17 Distribution plan generation unit 18 Calculation unit 19 Generation unit 20 Output unit 21 Output setting function unit 100, 200 Change management system 900 External system

Claims (5)

A POS system;
a prediction analysis unit which uses data on transactions in the POS system and data on each denomination in the change machine , sets explanatory variables as the number of deposits and withdrawals of each denomination for each time period, sets objective variables as the number of deposits and withdrawals of each denomination, and performs machine learning using store information, denominations of change, withdrawal and deposit information for each denomination, and purchased product information as learning data, creates a prediction model , and performs analysis by applying the generated number of deposits and withdrawals of each denomination for each time period to the prediction formula, thereby predicting the increase or decrease in change of each denomination for each target store during a target period;
A distribution calculation unit that uses the prediction results by the prediction analysis unit, inputs a road map, information on the location of a distribution office, and information on employees and delivery vehicles, and calculates the number of distributions of each denomination and the distribution cycle,
The dividend calculation unit is
classifying the target stores for each of the cash distribution cycles based on the prediction of the change increase or decrease for each denomination based on the prediction result ;
Calculating the distance to each target store when the location of the cash distribution office is set as the starting point according to the road map and the information on the location of the cash distribution office;
According to the prediction result, the road map, and the information on the location of the distribution office, for each distribution cycle, assign stores to the distribution office according to a predetermined distance from the distribution office and the number of stores, and list the stores by office;
When calculating a route for distributing money from the business establishment as a starting point for each distribution cycle for each list according to the information on the employees and delivery vehicles, information on the distribution date is matched with employee shift information and vehicle allocation information to generate a distribution schedule.
Change management system. The change management system according to claim 1, further comprising a setting unit that calculates a first threshold value, which is an upper limit of the change amount in the change machine for each of the target stores, and a second threshold value, which is a lower limit of the change amount, from information regarding the storage capacity of the change machine for each of the target stores and the prediction results. The change management system according to claim 1 or 2, wherein the prediction analysis unit takes into account data on customer attributes linked to data on transactions in the POS system when predicting the increase or decrease in change for each denomination at each target store during a target period. If a computer or system:
a first step of using data on transactions in the POS system and data on each denomination in the change machine , with explanatory variables being the number of deposits and withdrawals of each denomination for each time period and the number of deposits and withdrawals of each denomination for each time period, and assuming that the prediction model consists of a prediction formula, machine learning is carried out using store information, change denominations, withdrawal and deposit information for each denomination, and purchased product information as learning data to create a prediction model , and analyzing the generated number of deposits and withdrawals of each denomination for each time period by applying the prediction formula to the generated number of deposits and withdrawals of each denomination for each time period, thereby predicting the increase or decrease in change of each denomination for each target store during a target period;
a step of using the prediction results from the first step, inputting a road map, information on the location of the cash distribution office, and information on employees and delivery vehicles, classifying target stores for each cash distribution cycle based on a prediction of an increase or decrease in change for each denomination based on the prediction results , calculating a distance to each target store when the location of the cash distribution office is used as a starting point according to the road map and the information on the location of the cash distribution office, assigning stores to the cash distribution office for each cash distribution cycle according to a predetermined distance and number of stores in order of proximity to the cash distribution office according to the prediction results , the road map, and the information on the location of the cash distribution office, listing the stores by office, and calculating a route for distributing money starting from the office for each distribution cycle for each list according to the information on the employees and delivery vehicles, by matching information on the distribution date with employee shift information and vehicle allocation information, and calculating the number of changes to be distributed for each denomination and the distribution cycle. a first step of using data on transactions in the POS system and data on each denomination in the change machine , with explanatory variables being the number of deposits and withdrawals of each denomination for each time period and the number of deposits and withdrawals of each denomination for each time period, and assuming that the prediction model consists of a prediction formula, machine learning is carried out using store information, change denominations, withdrawal and deposit information for each denomination, and purchased product information as learning data to create a prediction model , and analyzing the generated number of deposits and withdrawals of each denomination for each time period by applying the prediction formula to the generated number of deposits and withdrawals of each denomination for each time period, thereby predicting the increase or decrease in change of each denomination for each target store during a target period;
a step of using the prediction results from the first step, inputting a road map, information on the location of the cash distribution office, and information on employees and delivery vehicles, classifying target stores for each cash distribution cycle based on a prediction of an increase or decrease in change for each denomination based on the prediction results , calculating a distance to each target store when the location of the cash distribution office is used as a starting point according to the road map and the information on the location of the cash distribution office, assigning stores to the cash distribution office for each cash distribution cycle according to a predetermined distance and number of stores in order of proximity to the cash distribution office according to the prediction results , the road map, and the information on the location of the cash distribution office, listing the stores by office, and calculating a route for distributing money starting from the office for each cash distribution cycle for each list according to the information on the employees and delivery vehicles, by matching information on the distribution date with employee shift information and vehicle allocation information, and calculating the amount of cash to be distributed for each denomination.