Sales promotion support systems, methods, and programs

Abstract

To provide a technique that enables distribution of electronic coupons taking the movement state of users into consideration.SOLUTION: A terminal device accumulates target user's past day movement history information that records movement history of a target user carrying the terminal device in past days. A sales promotion support device accumulates user movement history information that records movement history of the target user and other users in the past days. The terminal device records movement history of the target user up to the current time on the current day as target user's current day movement history information, uses the target user's current day movement history information and the target user's past day movement history information to generate first movement information regarding a series of movements including a predicted movement of the target user after the current time, and transmits the first movement information to the sales promotion support device. The sales promotion support device determines an electronic coupon to be distributed to the target user based on the first movement information and the user movement history information, and transmits the electronic coupon to the terminal device.SELECTED DRAWING: Figure 1

Description

【Technical Field】 【0001】 This disclosure relates to a technology for assisting sales promotion using electronic coupons. 【Background Art】 【0002】 Patent Document 1 discloses an information providing system for assisting sales promotion. The information providing system of Patent Document 1 notifies a terminal device of a user who intends to purchase a product of stores where the product can be purchased at a low price or stores where shopping can be done efficiently using the Internet or the like. Even if the user does not perform a registration process, information on various stores can be obtained. On the store side, an increase in the number of customers and sales due to the transmission of information can be expected. 【Prior Art Documents】 【Patent Documents】 【0003】 【Patent Document 1】 Japanese Patent Application Laid-Open No. 2012-190352 【Summary of the Invention】 【Problems to be Solved by the Invention】 【0004】 For example, even if product or store information is provided to a user in a state where there is no freedom to choose actions, such as when moving while about to go to work, high effects cannot be expected. However, in the technology of Patent Document 1, the moving state, such as in which state in a series of movements from leaving home to going to a destination and then returning home after completing the purpose, is not considered. Therefore, the information providing system of Patent Document 1 was not necessarily able to provide appropriate information. 【0005】 One of the purposes included in this disclosure is to provide technology that enables the distribution of electronic coupons while taking into account the user's movement status. For example, it is to provide technology that enables the selective distribution of electronic coupons to users who belong to segments that are expected to have a high probability of using electronic coupons, such as users who are on their way home after completing their purpose, or users who are taking different actions than usual due to business trips or detours. [Means for solving the problem] 【0006】 A sales promotion support system according to one aspect of the present disclosure includes a terminal device that stores target user past-day movement history information recording the movement history of a target user carrying the terminal device in the past day, a sales promotion support device that stores user movement history information recording the movement history of the target user and other users in the past day, a terminal device that records the movement history of the target user up to the present time of the day as target user current-day movement history information, a first movement information relating to a series of movements including the target user's predicted movements from the present time onward using the target user current-day movement history information and the target user past-day movement history information, and transmits it to the sales promotion support device, and a sales promotion support device that determines an electronic coupon to distribute to the target user based on the first movement information and the user movement history information, and transmits the electronic coupon to the terminal device. [Effects of the Invention] 【0007】 According to one aspect of this disclosure, it becomes possible to distribute electronic coupons that take into account the user's movement status. [Brief explanation of the drawing] 【0008】 [Figure 1] This is a block diagram of a sales promotion support system. [Figure 2] This is a flowchart for sales promotion support processing. [Figure 3] This diagram shows the structure of the user behavior model. [Figure 4] This figure shows the relationship between congestion rate, discount rate, and the probability of store visit in a user behavior model. [Figure 5] This is a flowchart of the coupon request processing and coupon distribution processing. [Figure 6] This is a flowchart for the process of generating movement feature information. [Figure 7] This is a flowchart of the segment determination process. [Figure 8] This is a diagram showing the structure of the segment classification model. [Modes for carrying out the invention] 【0009】 Embodiments of the present invention will be described below with reference to the drawings. 【0010】 Figure 1 is a block diagram of the sales promotion support system. 【0011】 The sales promotion support system 1 is an information processing system that distributes electronic coupons to users that offer benefits usable at stores. The sales promotion support system 1 includes a sales promotion support device 10, a terminal device 20, a travel demand forecasting system 32, and a travel forecasting system 33. The sales promotion support device 10, the terminal device 20, the travel demand forecasting system 32, and the travel forecasting system 33 are connected to a communication network 31. The communication network 31 includes, for example, the internet, an intranet, a wired or wireless LAN (Local Area Network), or a WAN (Wide Area Network). 【0012】 The terminal device 20 is a computer composed of portable hardware such as a smartphone or tablet terminal that the user carries and receives electronic coupons from. The terminal device 20 has a communication unit 21, an arithmetic processing unit 22, an input unit 23, a display unit 24, and a storage unit 25. The storage unit 25 stores a coupon request program 251, a movement characteristic information generation program 252, daily movement history information 253, past movement history information 254, and coupon request area information 255. 【0013】 The arithmetic processing unit 22 is a processor such as a CPU (Central Processing Unit) or GPU (Graphics Processing Unit). The arithmetic processing unit 22 uses the current day's movement history information 253, past movement history information 254, and coupon request area information 255 to execute the coupon request program 251 and the movement characteristic information generation program 252. The coupon request program 251 is a software program that causes the terminal device 20 to execute a process (coupon request process) to request an electronic coupon from the sales promotion support device 10. The movement characteristic information generation program 252 is a software program that causes the terminal device 20 to execute a process (movement characteristic information generation process) to calculate various characteristics of the user's movement on that day. In this specification, processes that are realized by the arithmetic processing unit 22 executing a software program in the terminal device 20 may be described as being executed by that software program. The coupon request process and the movement characteristic information generation process will be described later. 【0014】 The daily travel history information 253 is information that records the travel history of the user's terminal device 20 up to the current time on the current day. The past travel history information 254 is information that records the travel history of the user's terminal device 20 on past days included in a predetermined past period. The predetermined past period is, for example, one month. This travel history is time-series data recorded by, for example, sampling the position of the terminal device 20 at predetermined time intervals (e.g., 5 minutes), identifying an area with a predetermined resolution from the measured coordinates, and associating the area ID representing the area with the time the position was measured. The past travel history information 254 may be compressed and stored using run-length encoding or the like. Also, when the user is traveling by train, the recording of the position between stations may be omitted and not recorded in the daily travel history information 253 and the past travel history information 254. In that case, in order to make the total data length of the current day's travel history information 253 and the past travel history information 254 the same as the total data length of the current day's travel history information 253 and the past travel history information 254 when there are no omissions, the data for the omitted time may be filled with an exceptional value. 【0015】 The coupon request area information 255 is information that records the area ID indicating the coupon request area of the user. The coupon request area is an area where the user's stay is expected among the areas (target areas) where sales promotion measures are to be taken in the sales promotion support system 1. The coupon request area is an area where when the user's terminal device 20 enters that area, the sales promotion support device 10 is requested for an electronic coupon. For example, it may be possible to set the entire area around a certain station as one target area. For example, a target area where the user changes means of public transportation during movement is registered as the user's coupon request area. 【0016】 Here, an example of registering a target area where the user changes means of public transportation as the coupon request area has been shown, but other examples are also possible. For example, it may be possible to register a target area where the time the user has stayed exceeds a predetermined time as the coupon request area when the user exits that target area. 【0017】 Also, for the areas around facilities where many users gather, such as city halls, hospitals, department stores, etc., they may be registered in advance as common coupon request areas for all users. That is, the registered coupon request areas do not necessarily have to be mutually exclusive. For example, another coupon request area may be included within a certain coupon request area. 【0018】 The communication unit 21 is a communication device that enables communication via the communication network 31. The input unit 23 is an input device such as a touch panel or operation buttons through which the user operates to input information. The display unit 24 is a display device that displays images and text. 【0019】 The sales promotion support device 10 is a physical server composed of hardware computers that transmit electronic coupons to the terminal device 20, or a virtual server composed of computers virtually configured on the cloud. The sales promotion support device 10 has an arithmetic processing unit 11, a storage unit 12, an input unit 13, a display unit 14, and a communication unit 15. In the storage unit 12, a sales promotion support program 121, a congestion prediction program 122, an optimal measure formulation program 123, a segment determination program 124, a coupon distribution program 125, optimization condition data 126, and a segment classification model 127 are stored. Further, the sales promotion support device 10 stores coupon usage result data 16, a user behavior model 18, and a movement history database 17. 【0020】 The arithmetic processing unit 11 executes the sales promotion support program 121, the congestion prediction program 122, the optimal measure formulation program 123, the segment determination program 124, and the coupon distribution program 125 using the optimization condition data 126, the segment classification model 127, the coupon usage result data 16, the user behavior model 18, and the movement history database 17. In this specification, the processing realized by the arithmetic processing unit 11 executing a software program in the sales promotion support device 10 may be described as being executed by the software program. The processing of the sales promotion support program 121, the congestion prediction program 122, the optimal measure formulation program 123, the segment determination program 124, and the coupon distribution program 125 will be described later. 【0021】 Optimization condition data 126 is data that defines the conditions used to optimize the distribution plan, which determines how many electronic coupons to distribute to each segment of users. Details of the segment classification will be described later. Optimization condition data 126 includes data that is registered in advance and data that is registered when planning sales promotion measures for a certain period. The data that is registered in advance includes store information, including the number of seats in the store and the number of customers during periods when no sales promotion measures are being implemented. The data that is registered when planning sales promotion measures for a certain period includes the target period and the sales promotion budget, which is the amount of money allocated for distributing electronic coupons during the target period. 【0022】 Segment classification model 127 is a machine learning model (a model created in advance using machine learning) for classifying users into segments. Segment classification model 127 identifies the segment to which a user belongs based on attribute information indicating the user's attributes, performance information indicating the user's visits, purchases, and coupon usage, reference time information indicating the time used as the basis for determination, travel information regarding the user's movements on the day, and special day level information indicating the degree of unusualness of the user's movements on the day. User attribute information includes, for example, age, gender, occupation, hobbies, and preferences. 【0023】 The coupon usage data 16 is data that records each user's use of electronic coupons. For example, the coupon usage data 16 records the use of electronic coupons, the congestion status (congestion rate) indicating the degree of congestion on public transportation when the electronic coupon was used, and the distribution record of the electronic coupon at the time of use, all in correspondence with each other. 【0024】 The movement history database 17 is a database that records the past movement history of users in the target area of ​​a sales promotion campaign, linked to the user's attributes and the type of day. The type of day is information that indicates the type of day, such as the day of the week or whether it is a weekday or a holiday. 【0025】 The user behavior model 18 is set up for each segment and is a machine learning model (a model created in advance using machine learning) that calculates the probability of a user visiting a store based on the content of the benefits offered by the electronic coupons distributed within that segment. 【0026】 The sales promotion support program 121 is a software program that causes the sales promotion support device 10 to execute the overall sales promotion support process (sales promotion support process). The congestion prediction program 122 is a software program that causes the sales promotion support device 10 to execute the process (congestion prediction process) of predicting the degree of congestion of public transportation available for the user to travel. 【0027】 The Optimal Strategy Planning Program 123 is a software program that causes the sales promotion support device 10 to execute a process (distribution plan creation process) to create a distribution plan with the best cost-effectiveness. The distribution plan creation process is the process of creating a distribution plan that optimizes cost-effectiveness using the user behavior model 18 under the conditions specified in the optimization condition data 126. The segment determination program 124 is a software program that causes the sales promotion support device 10 to execute a process (segment determination process) to identify the user's segment. The segment determination process is the process of identifying the user's segment using the segment classification model 127. The coupon distribution program 125 is a software program that causes the sales promotion support device 10 to execute a process (coupon distribution process) to distribute electronic coupons to users. The coupon distribution process is the process of receiving a coupon distribution request from the terminal device 20 and transmitting the electronic coupon. 【0028】 Figure 2 is a flowchart of the sales promotion support process. 【0029】 First, the sales promotion support program 121 in the sales promotion support device 10 sets the target period and the budget amount to be allocated to the electronic coupons to be distributed during that period as optimization conditions (step S101). In addition, the optimization conditions also include store information, including the number of seats in the store and the number of customers visiting the store during periods when no sales promotion measures are being implemented. 【0030】 Next, the sales promotion support program 121 acquires demand forecast data from the mobility demand forecasting system 32 (step S102), and calls the congestion forecasting program 122 to execute congestion forecasting processing (step S103). 【0031】 In the congestion prediction process, the congestion prediction program 122 performs a passenger flow simulation based on the operation prediction data obtained from the operation prediction system 33, which shows the predicted operation of public transportation, and the demand prediction data obtained from the travel demand prediction system 32, and calculates the predicted degree of congestion at public transportation stations or for each service. Public transportation is, for example, a railway, in which case the congestion prediction program 122 calculates the degree of congestion at each railway station and each railway train. Note that while railways are shown here as an example of public transportation, public transportation is not limited to railways. Other examples of public transportation include route buses and LRT (Light Rail Transit). In the case of route buses, the congestion prediction program 122 calculates the degree of congestion at each bus stop and for each bus service. 【0032】 Next, the sales promotion support program 121 calls the optimal strategy planning program 123 to execute the distribution plan creation process (step S104). In the distribution plan creation process, the optimal strategy planning program 123 creates a distribution plan that optimizes cost-effectiveness using the user behavior model 18 under the conditions specified in the optimization condition data 126. 【0033】 Figure 3 shows the configuration of the user behavior model. Figure 4 shows the relationship between congestion rate, discount rate, and store visit probability in the user behavior model. As shown in Figure 3, the user behavior model 18 is set up for each segment, and in that segment, the discount rate of the electronic coupon to be distributed and the congestion rate of public transportation are inputs, and the store visit probability, which is the probability that a user will come to the store, is output. As shown in Figure 4, in the relationship between the congestion rate of public transportation and the discount rate of electronic coupons and the store visit probability in the user behavior model, a higher discount rate leads to a higher store visit probability, and a higher congestion rate leads to a relatively higher store visit probability. 【0034】 The distribution plan specifies how many electronic coupons to distribute to each segment. For example, electronic coupons offering various discount rates may be distributed to users randomly selected with a predetermined distribution probability. The cost is the budgeted amount allocated for distributing the electronic coupons. Alternatively, the cost may be calculated by multiplying the expected number of customers who will visit the store as a result of the distributed electronic coupons, the average customer spending, and the discount rate of the distributed electronic coupons. The effect is the increase in the number of customers visiting the store, the store's sales, or the increase in those figures. 【0035】 For example, the problem could be formulated as a mixed-integer programming problem, and the goal could be to find a distribution plan that can achieve the highest possible sales within a given promotional budget for the entire target period. The objective function and constraints could be created considering factors such as the acceptable level of store congestion, average dwell time, and average customer spending. The acceptable level of store congestion, average dwell time, and average customer spending could be set by the operator of the sales promotion support system 1 based on specifications from the store, etc. Setting a high acceptable level of store congestion would increase the waiting time before customers are seated. For example, based on the congestion forecast results for public transportation at each time of day for each day, one could select which segment's user behavior model 18 to use, input the discount rate and congestion rate into the selected user behavior model 18 to calculate the probability of a customer visiting the store, and then calculate the number of customers who will come to the store from that probability. Alternatively, sales could be calculated by further multiplying the number of customers who come to the store by the average customer spending. In this case, it is also possible to optimize multiple types of businesses simultaneously, such as restaurants and cafes, where the number of customers visiting may mutually influence each other, and adjust the supply and demand balance. 【0036】 Once the processing for the target period up to this point is complete, the sales promotion support program 121 will then perform daily processing (same-day processing) within the target period. 【0037】 In the same-day processing, first, the sales promotion support program 121 instructs the coupon distribution program 125 to execute the coupon distribution process (step S105). As described above, the coupon distribution process is a process that works in conjunction with the coupon request process in the terminal device 20. 【0038】 Figure 5 is a flowchart of the coupon request process and the coupon distribution process. 【0039】 The coupon request process begins when the terminal device 20 is powered on or the dedicated application is launched. First, the coupon request program 251 starts registering the day's travel history (step S201). The travel history registration is performed periodically at predetermined time intervals. 【0040】 Furthermore, the coupon request program 251 determines whether the terminal device 20 has entered the coupon request area based on the terminal device 20's current location information and the coupon request area information 255 (step S202). If the terminal device 20's current location moves from outside to inside the coupon request area registered in the coupon request area information 255, it is determined that the terminal device 20 has entered the coupon request area. If the terminal device 20 has entered the coupon request area, the coupon request program 251 calls the movement feature information generation program 252 to execute the movement feature information generation process (step S203). 【0041】 Figure 6 is a flowchart of the process for generating movement feature information. 【0042】 The movement feature information generation process calculates the characteristics of a user's movement on a given day and generates movement feature information. This movement feature information includes the movement information for that day (also called "first movement information" or "same-day movement information") and the singularity-level information for that day. 【0043】 First, the movement feature information generation program 252 performs the process of estimating movement information for the day (step S401). 【0044】 In the daily movement information estimation process, the movement feature information generation program 252 refers to the past movement history information 254 and generates the user's daily movement information, which shows the movement for the day, including movement after the estimated current time, based on the movement history up to the current time shown in the daily movement history information 253. For estimating movement after the current time, known techniques such as the nearest neighbor method can be used. Since the movement history is time-series data, when using the nearest neighbor method, the calculation of the similarity between vectors may use a calculation method such as elastic matching (DP (Dynamic Programming) matching, etc.) to absorb fluctuations in the time direction. If there is not enough data in the daily movement history information 253 to match with the movement history included in the past movement history information 254, or if there is no data in the past movement history information 254 whose similarity to the daily movement history information 253 is above a predetermined threshold, the movement history for the time range included in the daily movement history information 253 may be copied as is to the daily movement information (first movement information), and an exception value indicating that estimation is not possible for other times may be registered. 【0045】 Next, the movement feature information generation program 252 performs a stop point sequence extraction process (step S402). The movement feature information generation program 252 uses the movement history up to the current time of the day to extract an area ID sequence from the day's movement history information 253, which includes the station where the user boards or alights from a train or transfers (public transportation hubs (transportation hubs)) and the location where the user is expected to stay for a predetermined time or longer. The stop points are expected to be transit points or destinations of the user's movement on the day, and the stop point sequence is expected to be a sequence of locations that represent the characteristics of the user's behavior on the day. Multiple locations included in the same area may be combined into a single stop point. Furthermore, the movement feature information generation program 252 also performs a stop point sequence extraction process for each day included in the past movement history information 254 and extracts a similar area ID sequence. 【0046】 Next, the movement feature information generation program 252 executes the point sequence similarity calculation process (step S403). In the point sequence similarity calculation process, the movement feature information generation program 252 treats the area ID columns extracted in step S402 as point sequences and calculates the similarity of one point sequence to other point sequences. The similarity of one point sequence to other point sequences is the sum of the number of elements in one point sequence that do not match the elements in each of the other point sequences. Here, this sum is called the number of differing elements. Note that the number of differing elements may be normalized according to the length of the point sequence. 【0047】 Next, the movement feature information generation program 252 executes the process of calculating the unusual day level (step S404). The "unusual day level" is an index value that represents the degree to which a day is "different from usual" for the traveler under consideration, i.e., the degree of the unusualness. A high unusual day level suggests a high probability that the traveler will stop at a store or purchase goods during their journey. For example, the unusual day level may be calculated as: Unusual Day Level = (Number of differing elements corresponding to the travel history for the day) / (Average value of the number of differing elements corresponding to the travel history for past days). If the travel on the day differs from that of past days, the numerator of the above formula becomes larger, and the unusual day level becomes a high value. Also, if there is a large variation in the way traveled in past days, the denominator of the above formula becomes larger, and the unusual day level becomes a relatively low value. Here, an example of calculating the unusual day level using point sequence similarity is shown, but it is not limited to this. As another example, the unusual day level may be calculated using the similarity of the travel history itself. By calculating using the similarity of the travel history itself, it is possible to reflect in the unusual day level that work was completed earlier on that day than on past days. Thus, actions that are reflected at the level of a special day are not limited to simply going to a different place on that day than on previous days. 【0048】 Returning to Figure 5, in the terminal device 20, the coupon request program 251 sends a coupon distribution request to the sales promotion support device 10 (step S204). The coupon distribution request is a message signal requesting the distribution of an electronic coupon. The coupon distribution request includes the coupon request area where the user is staying, the current time, the user's attributes, the user's performance information, the user's movement information for the day (today's movement information), and the special day level for the user for that day. When the sales promotion support device 10 receives this coupon distribution request, the coupon distribution program 125 starts the coupon distribution process. 【0049】 First, when the coupon distribution program 125 receives this coupon distribution request (step S301), it calls the segment determination program 124 to execute the segment determination process (step S302). 【0050】 Figure 7 is a flowchart of the segment determination process. 【0051】 First, the segment determination program 124 obtains information about the user from the coupon distribution request (step S501). The user information includes the coupon request area where the user is currently located (area of ​​interest), the current time, the user's attributes, the user's performance information, the user's movement information for the day (movement information for the day), and the special day level for the user for that day. If the coupon request area obtained from the coupon distribution request is included in other coupon request areas, the segment determination program 124 may treat the coupon request area that covers a wider area as the area of ​​interest. 【0052】 Next, the segment determination program 124 performs movement pattern estimation processing (step S502). The segment determination program 124 refers to the movement history database 17 and, based on the current time and the movement information for the day, estimates the user's movement around the area of ​​interest for the day and outputs it as a movement pattern (second movement information). For movement estimation, known techniques such as the nearest neighbor method can be used. When using the nearest neighbor method, the calculation of the similarity between vectors may use calculation methods such as elastic matching (DP matching, etc.) to absorb fluctuations in the time direction. In order to estimate movement around the area of ​​interest, the range from just before entering the area of ​​interest to just after leaving the area of ​​interest is targeted for analysis. For other locations in the movement history database 17, the area ID can be replaced with an exception value such as "outside area" and processed accordingly. If there is not enough data in the movement information for the day to match with the movement history of past users registered in the movement history database 17, it is not possible to estimate movement around the area of ​​interest, so an exception value such as "unable to estimate" is output. 【0053】 The terminal device 20 performs a daily movement information estimation process that matches the user's movement history for the day with their past movement history. However, the movement pattern estimation process in step S502 of the sales promotion support device 10 estimates the user's movement pattern by utilizing the movement history of other users. Therefore, even if the user has never visited the area of ​​interest in the past, it is possible to estimate the user's movement pattern in that area of ​​interest. 【0054】 The scope of matching in the movement history database 17 may or may not include the day type and / or user attributes. Including them leads to the estimation of behavior such as, "If the user is in their 40s, went to the prefectural park on Saturday, and is still within the area of ​​interest around 8 PM, then they will likely stay overnight near the area of ​​interest." Alternatively, a sequence of stopping points can be extracted using a method similar to the method used in the movement feature information generation process in step S203, and by considering the time used as the basis for judgment, a label such as "return trip from area ID XX" can be assigned to the sequence of stopping points to give it meaning. 【0055】 Next, the segment determination program 124 performs segment classification processing using the segment classification model 127 (step S503). 【0056】 Figure 8 shows the structure of the segment classification model. 【0057】 By inputting the user's attribute information, performance information, reference time information, movement pattern information, and special day level information into the segment classification model 127, a segment ID indicating the segment to which the user belongs can be obtained. In the same-day processing, the current time is given as the reference time information, so even the same user may be classified into different segments depending on the current time. The segment classification model 127 is to be created in advance using machine learning. A known model such as a decision tree or neural network can be used, and training can be performed using a known machine learning framework under predetermined hyperparameters. 【0058】 Returning to Figure 5, once the segment determination process is complete, the coupon distribution program 125 executes the coupon distribution determination process (step S303). The coupon distribution determination process determines the content of the electronic coupon to be sent to the user. For example, if the distribution plan stipulates that electronic coupons offering various discount rates be distributed to users at predetermined distribution probabilities, the coupon distribution program 125 should decide to send an electronic coupon with a randomly determined discount rate to the user based on the predetermined distribution probability. 【0059】 Furthermore, the coupon distribution program 125 may adjust the discount rate and / or distribution probability of the electronic coupons according to the actual number of users visiting the store or the level of congestion in the store, so that the number of visitors or the probability of visitors is roughly the same as the value assumed when the distribution plan was created. 【0060】 Next, the coupon distribution program 125 transmits the coupon distribution result to the user's terminal device 20 (step S304). The coupon distribution result is information indicating the content of the electronic coupon to be distributed to the user. Then, the coupon distribution program 125 determines whether or not the coupon distribution end time has been reached (step S305). The coupon distribution end time is the time when the target period ends. The coupon distribution program 125 repeats the processes from steps S301 to S305 described above until the coupon distribution end time is reached. 【0061】 When the coupon request program 251 in the terminal device 20 receives the coupon distribution result from the sales promotion support device 10 (step S205), it displays the distributed electronic coupon on the display unit 24 (step S206). If no electronic coupon is distributed, the coupon request program 251 displays nothing. 【0062】 After completing the processing in step S206, or after determining in step S202 that the user has not entered the coupon request area, the coupon request program 251 determines whether the user's terminal device 20 has left the traffic hub area (step S207). 【0063】 If terminal device 20 has left the transportation hub area, coupon request program 251 checks whether that transportation hub area is registered as a coupon request area, and registers it if it is not (step S208). Coupon request program 251 also transmits information about the movement history from entering to leaving the coupon request area to sales promotion support device 10. This information is registered in the movement history database 17 of sales promotion support device 10. Alternatively, coupon request program 251 may transmit information about the movement history for the day to sales promotion support device 10, and sales promotion support device 10 may extract the portion of the information received from entering to leaving the coupon request area and register it in the movement history database 17. 【0064】 In terminal device 20, the coupon request program 251 repeats the processes from steps S201 to S208 described above until it receives a termination instruction (step S209). The termination instruction is, for example, the termination of a dedicated application. 【0065】 Returning to Figure 2, the sales promotion support program 121 executes the coupon usage data update process (step S106). The coupon usage data update process is the process of recording information regarding the use of electronic coupons on that day in the coupon usage data 16. The sales promotion support program 121 records in coupon usage data 16 the usage of electronic coupons, the congestion status (congestion rate) indicating the degree of congestion on public transportation when the electronic coupon was used, and the actual distribution of electronic coupons when the electronic coupon was used. 【0066】 Next, the sales promotion support program 121 executes a user behavior model update process (step S107). In the user behavior model update process, the sales promotion support program 121 updates the user behavior model for each segment based on the actual distribution of electronic coupons, the coupon usage data 16, and the number of users who actually visited the store. 【0067】 Next, the sales promotion support program 121 determines whether the current day is the last day of the target period (step S108). If the current day is the last day, the sales promotion support program 121 terminates the sales promotion support process. If the current day is not the last day, the sales promotion support program 121 updates the remaining number of days in the target period and the remaining sales promotion budget under the optimization conditions (step S109), and returns to step S102 to repeat the subsequent processing. 【0068】 The above-described embodiment includes the following matters. However, the matters included in this embodiment are not limited to those listed below. 【0069】 (Item 1) A sales promotion support system that distributes electronic coupons to users that offer benefits usable at stores, A terminal device carried by the user and used to receive the electronic coupon, A sales promotion support device that transmits the electronic coupon to the terminal device, It has, The terminal device stores user past-day travel history information, which records the travel history of the target user carrying the terminal device on past days. The sales promotion support device stores user movement history information that records the movement history of the target user and other users on past days. The aforementioned terminal device The movement history of the aforementioned user up to the current time on that day is recorded as the user's movement history information for that day. Using the target user's current day travel history information and the target user's past day travel history information, first travel information relating to a series of movements including the target user's predicted movements from the current time onward is generated and transmitted to the sales promotion support device. The aforementioned sales promotion support device Based on the first movement information and the user movement history information, the electronic coupon to be distributed to the target user is determined. The electronic coupon is transmitted to the terminal device. Sales promotion support system. 【0070】 According to this system, the terminal device stores the travel history information of the target user carrying it for past days, the sales promotion support device stores the travel history information of the target user and other users, the terminal device records the target user's travel history information for the current day, and uses the target user's travel history information for the current day and the target user's travel history information for past days to generate first travel information regarding a series of movements, including predicted movements from the current time onward. The sales promotion support device then determines which electronic coupons to distribute to the target user based on the first travel information and the travel history information of users, including users other than the target user, that is stored in the device. This makes it possible to distribute electronic coupons to the target user even when the target user is visiting an area for the first time, taking into account the travel status of the target user and other users. 【0071】 (Item 2) In the sales promotion support system described in item 1, The aforementioned sales promotion support device is Using the first movement information and the user movement history information, second movement information is generated relating to a series of movements of the target user, including predicted movements from the current time onward. Based on the attributes of the target user, the current time used as the basis for determination, and the second movement information, the segment to which the target user belongs is determined. Based on the segment to which the target user belongs, the electronic coupon to be distributed to the target user is determined. 【0072】 According to this, the sales promotion support device generates second movement information that shows the estimated movement of the target user using the history of other users, and uses this second movement information to determine the user's segment. Therefore, even if the target user is visiting an area for the first time, it becomes possible to distribute electronic coupons with content that is effective for segments that take into account the target user's movement status (whether or not they have stopped by their destination) predicted from the movement of other users, to the target user's terminal device. 【0073】 (Item 3) In the sales promotion support system described in item 1, The aforementioned sales promotion support device stores user movement history information in the target area of ​​the sales promotion measures, The terminal device stores, based on the movement history of the target user, the areas within the target area where the target user is expected to stay as coupon request areas. When the terminal device enters the coupon request area, it transmits a coupon distribution request including the first movement information to the sales promotion support device. When the sales promotion support device receives the coupon distribution request, it transmits an electronic coupon to be distributed to the target user to the terminal device, based on the first movement information and the user movement history information. 【0074】 According to this system, when a target user's terminal device enters the sales promotion area, a coupon distribution request is sent, and the sales promotion support device distributes electronic coupons accordingly, making it possible to distribute electronic coupons at an effective time. 【0075】 (Item 4) In the sales promotion support system described in item 1, The aforementioned sales promotion support device is A segment classification model is used to determine the segment to which a user belongs based on attribute information indicating the user's attributes, reference time information indicating the time used as the basis for determination, and travel information regarding the user's movements on that day. For each segment, a user behavior model is pre-stored that represents the probability of a user visiting the store, which is the probability that the user will come to the store according to the content of the benefits of the electronic coupon distributed. Using the aforementioned user behavior model, a distribution plan is created that defines the content of the electronic coupon benefits to be distributed to each segment in order to maximize the number of users who come to the store within a predetermined sales promotion budget. Using the aforementioned segment classification model, the segment to which the target user belongs is determined based on the target user's attributes, the current time used as the basis for determination, and the target user's movement information. Based on the segment to which the target user belongs and the distribution plan, the electronic coupons to be distributed to the target user are determined. 【0076】 (Item 5) In the sales promotion support system described in item 4, The user behavior model is a model that represents the probability of visiting the store for each segment, in accordance with the content of the benefits of the electronic coupon and the congestion status, which indicates the degree of congestion of public transportation available to the user for travel. The aforementioned sales promotion support device is Predicting the aforementioned congestion situation, Using the user behavior model described above, the distribution plan is created based on the predicted congestion status. 【0077】 According to this, it becomes possible to create distribution plans that reflect the behavioral trends of users in each segment according to the congestion level of public transportation, enabling the effective distribution of electronic coupons according to the traffic congestion situation. 【0078】 (Item 6) In the sales promotion support system described in item 1, The aforementioned sales promotion support device A segment classification model is used to determine the segment to which a user belongs based on attribute information indicating the user's attributes, reference time information indicating the time used as the basis for determination, and travel information regarding the user's movements on that day. For each segment, a user behavior model is pre-stored that represents the probability of a user visiting the store, which is the probability that the user will come to the store according to the content of the benefits of the electronic coupon distributed. Using the aforementioned user behavior model, a distribution plan is created that defines the content of the electronic coupon benefits to be distributed to each segment in order to maximize the number of users who come to the store within a predetermined sales promotion budget. The terminal device stores user past-day travel history information, which records the travel history of the target user carrying the terminal device on past days. The aforementioned sales promotion support device stores user movement history information, which records the history of a user's movement in the target area of ​​the sales promotion measure over the past day. The terminal device stores, based on the movement history of the target user, the areas within the target area where the target user is expected to stay as coupon request areas. The aforementioned terminal device Record the user's movement history information for the day, which records the movement history of the aforementioned user up to the current time on that day. Upon entering the coupon request area, the system generates first movement information relating to a series of movements including the target user's predicted movements from the current time onward, using the target user's movement history information for the current day and the target user's movement history information for past days, and sends this information to the sales promotion support device. The aforementioned sales promotion support device Using the first movement information and the user movement history information, second movement information is generated relating to a predicted series of movements of the target user within the target area. Using the segment classification model described above, the segment to which the target user belongs is determined based on the attributes of the target user, the current time used as the criterion for determination, and the second movement information. Based on the segment to which the target user belongs and the distribution plan, the electronic coupon to be distributed to the target user is determined. The electronic coupon is transmitted to the terminal device. 【0079】 (Item 7) In the sales promotion support system described in item 6, The aforementioned segment classification model further determines the segment to which a user belongs based on an unusual day level that indicates the degree to which a user's movement on a given day differs from their usual movement. The terminal device calculates the level of unusual activity in the target user's movements on the current day using the target user's movement history information for the current day and the target user's movement history information for past days, and sends this information to the sales promotion support device. The sales promotion support device uses the segment classification model to determine the segment to which the target user belongs, based on the target user's attributes, the current time, the second movement information, and the special day level. 【0080】 According to this, by segmenting users based on the degree to which their travel differs from their usual routines, such as going to places they wouldn't normally visit, it becomes possible to distribute electronic coupons effectively, taking into account the target users' daily activities. 【0081】 The embodiments described above are illustrative for explaining the present invention and are not intended to limit the scope of the invention to those embodiments only. Those skilled in the art can implement the present invention in various other forms without departing from the scope of the invention. [Explanation of Symbols] 【0082】 1...Sales promotion support system, 10...Sales promotion support device, 11...Calculation processing unit, 12...Storage unit, 13...Input unit, 14...Display unit, 15...Communication unit, 16...Coupon usage data, 17...Movement history database, 18...User behavior model, 20...Terminal device, 21...Communication unit, 22...Calculation processing unit, 23...Input unit, 24...Display unit, 25...Storage unit, 31...Communication network, 32...Movement demand forecasting system, 33...Operation forecasting system, 121...Sales promotion support program, 122...Congestion forecasting program, 123...Optimal strategy planning program, 124...Segment determination program, 125...Coupon distribution program, 126...Optimization condition data, 127...Segment classification model, 251...Coupon request program, 252...Movement characteristic information generation program, 253...Current day's movement history information, 254...Past movement history information, 255...Coupon request area information

Claims

[Claim 1] A sales promotion support system that distributes electronic coupons to users that offer benefits usable at stores, A terminal device carried by the user and used to receive the electronic coupon, A sales promotion support device that transmits the electronic coupon to the terminal device, It has, The aforementioned sales promotion support device A segment classification model is used to determine the segment to which a user belongs based on attribute information indicating the user's attributes, reference time information indicating the time used as the basis for determination, and travel information regarding the user's movements on that day. For each segment, a user behavior model is pre-stored that represents the probability of a user visiting the store, which is the probability that the user will come to the store according to the content of the benefits of the electronic coupon distributed. Using the aforementioned user behavior model, a distribution plan is created that defines the content of the electronic coupon benefits to be distributed to each segment in order to maximize the number of users who come to the store within a predetermined sales promotion budget. The terminal device stores user past-day travel history information, which records the travel history of the target user carrying the terminal device on past days. The aforementioned sales promotion support device stores user movement history information, which records the history of a user's movement in the target area of ​​the sales promotion measure over the past day. The terminal device stores, based on the movement history of the target user, the areas within the target area where the target user is expected to stay as coupon request areas. The aforementioned terminal device Record the user's movement history information for the day, which records the movement history of the aforementioned user up to the current time on that day. Upon entering the coupon request area, the system generates first movement information relating to a series of movements including the target user's predicted movements from the current time onward, using the target user's movement history information for the current day and the target user's movement history information for past days, and sends this information to the sales promotion support device. The aforementioned sales promotion support device Using the first movement information and the user movement history information, second movement information relating to a predicted series of movements of the target user within the target area is generated. Using the segment classification model described above, the segment to which the target user belongs is determined based on the attributes of the target user, the current time used as the criterion for determination, and the second movement information. Based on the segment to which the target user belongs and the distribution plan, the electronic coupon to be distributed to the target user is determined. The electronic coupon is transmitted to the terminal device. Sales promotion support system. [Claim 2] The aforementioned segment classification model further determines the segment to which a user belongs based on an unusual day level that indicates the degree to which a user's movement on a given day differs from their usual movement. The terminal device calculates the level of unusual activity in the target user's movements on the current day using the target user's movement history information for the current day and the target user's movement history information for past days, and sends this information to the sales promotion support device. The sales promotion support device uses the segment classification model to determine the segment to which the target user belongs, based on the target user's attributes, current time, the second movement information, and the special day level. The sales promotion support system according to claim 1. [Claim 3] A sales promotion support method for distributing electronic coupons to a user, comprising: a terminal device that receives electronic coupons carried by the user and that provide benefits usable at a store; and a sales promotion support device that transmits the electronic coupons to the terminal device, The aforementioned sales promotion support device A segment classification model is used to determine the segment to which a user belongs based on attribute information indicating the user's attributes, reference time information indicating the time used as the basis for determination, and travel information regarding the user's movements on that day. For each segment, a user behavior model is pre-stored that represents the probability of a user visiting the store, which is the probability that the user will come to the store according to the content of the benefits of the electronic coupon distributed. Using the aforementioned user behavior model, a distribution plan is created that defines the content of the electronic coupon benefits to be distributed to each segment in order to maximize the number of users who come to the store within a predetermined sales promotion budget. The terminal device stores user past-day travel history information, which records the travel history of the target user carrying the terminal device on past days. The aforementioned sales promotion support device stores user movement history information, which records the history of a user's movement in the target area of ​​the sales promotion measure over the past day. The terminal device stores, based on the movement history of the target user, the areas within the target area where the target user is expected to stay as coupon request areas. The aforementioned terminal device Record the user's movement history information for the day, which records the movement history of the aforementioned user up to the current time on that day. Upon entering the coupon request area, the system generates first movement information relating to a series of movements including the target user's predicted movements from the current time onward, using the target user's movement history information for the current day and the target user's movement history information for past days, and sends this information to the sales promotion support device. The aforementioned sales promotion support device Using the first movement information and the user movement history information, second movement information relating to a predicted series of movements of the target user within the target area is generated. Using the segment classification model described above, the segment to which the target user belongs is determined based on the attributes of the target user, the current time used as the criterion for determination, and the second movement information. Based on the segment to which the target user belongs and the distribution plan, the electronic coupon to be distributed to the target user is determined. The electronic coupon is transmitted to the terminal device. Sales promotion support methods. [Claim 4] The segment classification model further determines the segment to which a user belongs based on an unusual day level that indicates the degree to which a user's movement on a given day differs from that user's usual movement, The terminal device calculates the level of unusual activity in the target user's movements on the current day using the target user's movement history information for the current day and the target user's movement history information for past days, and sends this information to the sales promotion support device. The sales promotion support device uses the segment classification model to determine the segment to which the target user belongs, based on the target user's attributes, current time, the second movement information, and the special day level. The sales promotion support method according to claim 3. [Claim 5] A sales promotion support program to be executed by a computer which is the sales promotion support device in a sales promotion support system that distributes electronic coupons to a user, comprising: a terminal device that receives electronic coupons that are carried by a user and provide benefits that can be used at a store; and a sales promotion support device that transmits the electronic coupons to the terminal device, the sales promotion support program to be executed by the computer which is the sales promotion support device A segment classification model is used to determine the segment to which a user belongs based on attribute information indicating the user's attributes, reference time information indicating the time used as the basis for determination, and travel information regarding the user's movements on that day. For each segment, a user behavior model is pre-stored that represents the probability of a user visiting the store, which is the probability that the user will come to the store according to the content of the benefits of the electronic coupon distributed. Using the aforementioned user behavior model, a distribution plan is created that defines the content of the electronic coupon benefits to be distributed to each segment in order to maximize the number of users who come to the store within a predetermined sales promotion budget. By accumulating user movement history information that records the history of users' movements in the target area of ​​sales promotion measures over the past day, The terminal device stores target user past day travel history information, which records the travel history of the target user carrying the terminal device in the past days; based on the target user's travel history, stores areas within the target area where the target user is expected to stay as coupon request areas; records target user current day travel history information, which records the target user's travel history up to the current time on that day; and when the target user enters the coupon request area, receives first travel information relating to a series of movements including the target user's predicted movements from the current time onward, which is generated using the target user current day travel history information and the target user past day travel history information. Using the first movement information and the user movement history information, second movement information relating to a predicted series of movements of the target user within the target area is generated. Using the segment classification model described above, the segment to which the target user belongs is determined based on the attributes of the target user, the current time used as the criterion for determination, and the second movement information. Based on the segment to which the target user belongs and the distribution plan, the electronic coupon to be distributed to the target user is determined. The electronic coupon is transmitted to the terminal device. A sales promotion support program to help achieve a goal. [Claim 6] The segment classification model further determines the segment to which a user belongs based on an unusual day level that indicates the degree to which the user's movement on that day differs from the user's usual movement, The terminal device receives the special day level of the target user's movement on the current day, calculated using the target user's movement history information for the current day and the target user's movement history information for past days. Using the segment classification model, the segment to which the target user belongs is determined based on the target user's attributes, current time, the second movement information, and the singular day level. A sales promotion support program according to claim 5, which causes a computer to perform the following actions.

Images