Method and server for predicting the total number of packages delivered to a destination

Abstract

To provide a method capable of reducing a package delivery load by efficiently using a home delivery box, and server for implementing the method.SOLUTION: A delivery method includes: causing an allocation server to predict the total number of packages to be delivered to a first home delivery box; in a case where there is a period when the predicted total number exceeds a capacity of the first home delivery box, causing the allocation server to select, as a first user, one of a plurality of users who have specified the first home delivery box as a delivery destination of a package in the period; transmitting a change request signal from the allocation server to a communication terminal of the first user, the change request signal for requesting a change in delivery date / time of the package or a change to a second home delivery box as a delivery destination of the package; and causing the allocation server to receive a response signal transmitted from the communication terminal to accepts or reject the change.SELECTED DRAWING: Figure 3A

Description

[Technical field]

[0001] One embodiment of the present invention relates to a delivery method for delivering packages and a server for implementing this delivery method. [Background technology]

[0002] There are cases where the user is not present when a package is delivered to the user. In this case, the package is brought into the delivery box and the user is notified of the delivery of the package, allowing the user to remove the package from the delivery box at any time. This reduces the burden and cost of redelivery, and also relieves the user from time constraints, as they no longer need to wait for delivery. It is also possible to prevent package theft and the leakage of personal information (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2020-077391 A Summary of the Invention [Problem to be solved by the invention]

[0004] An object of one embodiment of the present invention is to provide a novel method for delivering packages to users and a server for implementing this method. Alternatively, an object of one embodiment of the present invention is to provide a method capable of reducing the burden of package delivery by efficiently using delivery lockers and a server for implementing this method. [Means for solving the problem]

[0005] One embodiment of the present invention is a method for delivering packages, which includes: predicting, in an allocation server, the total number of packages to be delivered to a first delivery box; if there is a period during which the predicted total number exceeds the capacity of the first delivery box, selecting, as a first user, one of a plurality of users who have designated the first delivery box as the delivery destination of the packages during that period, by the allocation server; transmitting, to a communication terminal of the first user, a change request signal for requesting a change in the delivery date and time of the packages or a change in the delivery destination of the packages to a second delivery box; and receiving, in the allocation server, a response signal transmitted from the communication terminal to accept or reject the change.

[0006] One embodiment of the present invention is a server configured to predict the total number of packages to be delivered to a first delivery box, and when there is a period in which the predicted total number exceeds the capacity of the first delivery box, select one user as a first user from among multiple users who have specified the first delivery box as the delivery destination of the packages during that period, transmit a change request signal to a communication terminal of the first user to request a change in the delivery date and time of the packages or to change the delivery destination of the packages to a second delivery box, and receive a response signal transmitted from the communication terminal to accept or reject the change. [Brief description of the drawings]

[0007] [Figure 1A] FIG. 1 is a conceptual diagram illustrating a delivery method according to an embodiment of the present invention. [Figure 1B] 1 is a conceptual diagram of a system for implementing a delivery method according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a functional block diagram of an allocation server according to an embodiment of the present invention. [Figure 3A] 1 is a scheme illustrating an example of a delivery method according to an embodiment of the present invention. [Figure 3B] 1 is a scheme illustrating an example of a delivery method according to an embodiment of the present invention. [Figure 3C] 1 is a scheme illustrating an example of a delivery method according to an embodiment of the present invention. [Figure 4] 1 is a table showing an example of feature amounts used in a delivery method according to an embodiment of the present invention. [Figure 5A] FIG. 1 is a schematic diagram illustrating a delivery method according to an embodiment of the present invention. [Figure 5B] FIG. 1 is a schematic diagram illustrating a delivery method according to an embodiment of the present invention. [Figure 6] 1 is a flowchart for explaining a delivery method according to an embodiment of the present invention. [Figure 7] 11 is a table showing an example of user parameters used in a delivery method according to an embodiment of the present invention. [Figure 8] FIG. 2 is a schematic diagram showing signal transmission and reception in a delivery method according to an embodiment of the present invention. [Figure 9] FIG. 2 is a schematic diagram showing signal transmission and reception in a delivery method according to an embodiment of the present invention. [Figure 10] FIG. 2 is a schematic diagram showing signal transmission and reception in a delivery method according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] Hereinafter, each embodiment of the present invention will be described with reference to the drawings, etc. However, the present invention can be embodied in various forms without departing from the gist of the present invention, and should not be construed as being limited to the description of the embodiments exemplified below.

[0009] In the drawings, the width, thickness, shape, etc. of each part may be shown diagrammatically compared to the actual embodiment in order to make the explanation clearer, but it is merely an example and does not limit the interpretation of the present invention. In this specification and each figure, elements having the same function as those explained in the previous figures may be given the same reference numerals and duplicate explanations may be omitted. This reference numeral is used to collectively represent multiple identical or similar structures, and a hyphen and a natural number are added after the reference numeral to represent each of them individually.

[0010] Hereinafter, a package delivery method according to one embodiment of the present invention (hereinafter also referred to simply as this delivery method), an allocation server for realizing this delivery method, and a system capable of realizing this delivery will be described.

[0011] 1. Overview of Shipping Methods A conceptual diagram of this delivery method is shown in Fig. 1A. In this delivery method, a package including a product ordered by a customer (hereinafter, user) 100 is delivered by a so-called hold method. In other words, the package is not delivered to the door of the user 100, but is delivered by storing the package in a locker provided in a delivery box 190.

[0012] A user 100 orders a product from a store 180, which is a seller of the product, via various communication networks 104 such as a wide area network exemplified by the Internet, a telephone line, a fax line, or through face-to-face business transactions. The store 180 may be a real store such as a general retail store such as a supermarket or a department store, a liquor store, a dry cleaning shop, a restaurant or coffee shop, a fast food restaurant, a bookstore, a stationery store, or a drug store that handles cosmetics and medicines. The store 180 may also be a store that conducts electronic commerce on an electronic commerce platform that the store 180 has built itself, or an electronic commerce platform provided by a business entity (hereinafter, a platform provider entity) 160 that provides a platform for a fee or free of charge. Therefore, the store 180 does not necessarily have to be a real store. There are no restrictions on the products included in the package, and products belonging to any type of business are included. The products may also be rental products (rental shoes, rental costumes, rental bags, etc.) or samples (sample cosmetics, sample foods, etc.).

[0013] The entity that delivers the parcel (hereinafter, the delivery person) is the store 180 that received the order for the product, or the delivery company 170 that receives a delivery request from the store 180, the platform provider 160, or the allocation company 110 described below. Alternatively, the allocation company 110 may be in charge of delivery operations as the delivery person.

[0014] As described above, all delivered packages are left in the delivery box 190. Therefore, the delivery person does not need to carry the package to each residence, and does not need to check whether the user 100 is at home or not. Therefore, this delivery method places an extremely small burden on the delivery person, significantly reduces the time and cost required for delivery, and makes it possible to effectively utilize human resources in delivery operations.

[0015] In this delivery method, one or more of the lockers of the delivery box 190 designated by the user 100 are allocated to the package before the package is delivered. More specifically, the allocation company 110 selects one or more lockers suitable for the package from among the lockers of the delivery box 190 designated by the user 100. Furthermore, the package deliverer receives the allocation of the locker selected by the allocation company 110 via the communication network 104 before delivering the package, and then delivers the package. This makes it possible to secure a locker for storing the package before delivering it, and relieves the burden of redelivery that is forced because all the lockers are in use. When selecting a locker, the allocation company 110 periodically or at any time receives information on the usage status of the lockers of each delivery box 190 (hereinafter referred to as locker information) via the communication network 104 directly from the delivery box 190 or from a company that manages the delivery box 190 (hereinafter referred to as box management company 150). This makes it possible to promptly allocate a locker suitable for storing the package in response to the request of the deliverer.

[0016] As described in detail below, in this delivery method, the allocation business operator 110 executes a demand forecast for each home delivery box 190, and adjusts the delivery of parcels based on the demand forecast. Specifically, the allocation business operator 110 predicts the total number of parcels to be delivered to each home delivery box 190 using a learning model. If it is determined that there is a period (hereinafter, a busy period) in which the predicted total number of parcels exceeds the capacity of the home delivery box 190 (i.e., the total number of lockers in each home delivery box 190), one or more users 100 who have specified the home delivery box 190 as the delivery destination of parcels during the busy period are selected. The selection is performed according to user parameters described later. After that, a request is made to the selected user 100 to change the delivery date and time of the parcel, or to change the delivery destination of the parcel to another home delivery box (hereinafter, such requests to the user 100 are also referred to as promotions). This makes it possible to encourage some users 100 to use the home delivery box 190 while avoiding the busy period. Alternatively, it is possible to encourage the use of other home delivery boxes 190 that have spare capacity. As a result, it is possible to prevent the total number of parcels from exceeding the capacity of the parcel locker 190, so parcel deliveries are not concentrated in a certain period of time, and the parcel deliverers can secure lockers without fail. In addition, delivery work can be leveled out, which makes it possible to prevent unintended delivery delays.

[0017] If the user 100 accepts the promotion, the allocation company 110 may provide some kind of benefit (incentive) to the user 100. This motivates the user 100 to more actively accept the promotion from the allocation company 110, which results in smoother delivery coordination and delivery operations.

[0018] 2. Distribution System Configuration An example of the configuration of a delivery system 200 for realizing this delivery method is shown in FIG. 1B. The delivery system 200 includes an allocation server 112 managed by an allocation company 110, one or more delivery boxes 190, a box management server 152 used by a box management company 150 to manage the delivery boxes 190, and a delivery management server 172 used by a delivery company 170 to manage the delivery of packages. The delivery system 200 may further include a store server 182 used by a store 180 when placing an order from a user 100, and a platform server 162 used by a platform provider 160 to build a platform. The delivery system 200 may also include a user terminal 102, which is a communication terminal managed by the user 100. These components are connected to a communication network 104 and are in a state where they can communicate with each other.

[0019] When the allocation agent 110 is in charge of delivering packages as a delivery person, the allocation server 112 and the delivery management server 172 may be managed by the allocation agent 110, or these two servers may be integrated into one server that is managed by the allocation agent 110. The configuration of these servers will be explained below, but because the delivery management server 172, platform server 162, and store server 182 can be constructed using existing servers or communication terminals such as computers, explanations will be omitted.

[0020] (1) Delivery box and box management server Each delivery box 190 has multiple lockers and can be installed in any location, such as a common space in an apartment complex, a train station, a convenience store, a general retail store, an amusement facility, a parking lot, or a service area on a highway. The multiple delivery boxes 190 may all be managed by one business operator, or may be configured by multiple delivery boxes managed by multiple businesses that are independent of each other. The specifications of the multiple delivery boxes 190 may be the same as each other, or at least one of the specifications may be different from the others. The sizes of the multiple lockers provided in each delivery box 190 may be the same as each other, or at least one of the lockers may be different from the others. An identifier (box ID) for identifying each delivery box 190 is assigned to each delivery box 190, and an identifier (locker ID) for identifying each locker is assigned to each of the multiple lockers in each delivery box 190.

[0021] Although not shown, each locker of delivery box 190 may be equipped with a sensor or camera for grasping the presence or absence of luggage in the locker and the arrangement state thereof. Delivery box 190 may further be equipped with a sensor for detecting the opening and closing of each locker. For this reason, delivery box 190 may be configured so that, for example, when a locker is opened, a camera is used to obtain the internal arrangement state as an image (still image or video), and when the locker is opened or closed, the image data is transmitted to box management server 152 together with the time when the locker was opened or closed. This allows the time when the luggage was placed in the locker and the time when it was removed to be grasped on delivery box 190 or box management server 152.

[0022] The presence or absence of luggage in the locker detected by the sensor is transmitted as locker information to the box management server 152 periodically, in response to a request from the box management server 152, or in response to the operation of a sensor provided in the delivery box 190. Therefore, the box management server 152 can constantly know the locker information of the delivery box 190. The locker information is transmitted from the box management server 152 to the allocation server 112 periodically, or in response to a request from the allocation server 112. Therefore, the allocation server 112 can also constantly or timely know the locker information.

[0023] (2) Allocation Server A functional block diagram of the allocation server 112 according to one embodiment of the present invention is shown in Fig. 2. The allocation server 112 is a device having communication and calculation functions, and there are no restrictions on its configuration. For example, the allocation server 112 may include a control unit 114 that controls the entire allocation server 112, as well as a storage unit 122, a communication unit 124, an input unit 126, an output unit 128, and the like.

[0024] The input unit 126 is one of the interfaces for operating the allocation server 112, and is, for example, a keyboard, a mouse, a touch pen, a touch sensor, or the like. The output unit 128 is an interface for outputting information, and is, for example, a display device such as a liquid crystal display device or an electroluminescence display device. A part of the input unit 126 may be incorporated into the output unit 128 as a touch sensor. The storage unit 122 includes, for example, a main storage device such as a random access memory (RAM) or a read-only memory (ROM), and an external storage device such as a hard disk drive (HDD) or a flash memory. The storage unit 122 stores various information such as the calculation results and programs of the control unit 114, and also stores user parameters, which are information related to the user 100, and various information used for demand forecasting of the delivery box 190. The storage unit 122 further stores input data, teacher data, learning models, demand forecast results by the learning models, and the like, which are necessary for calculations for demand forecasting.

[0025] The communication unit 124 is a unit for communicating with the box management server 152, the platform server 162, the delivery management server 172, the store server 182, and the like via the communication network 104. The communication unit 124 communicates with, for example, the delivery box 190 or the box management server 152 to acquire locker information. Furthermore, the communication unit 124 may be configured to communicate with the store server 182 or the platform server 162 to acquire various information such as service information including whether or not each store 180 or platform provider 160 has offered a special offer related to a product, whether or not the user 100 has used the special offer, and the contents of the special offer.

[0026] The control unit 114 has a calculation device such as a central processing unit (CPU), reads out and executes necessary programs from the storage unit 122 to control the entire allocation server 112, and performs calculation processing according to the learning model. The control unit 114 can be divided into a demand forecasting unit 116, an adjustment unit 118, and a user management unit 120 as units that perform its main functions. Note that the control unit 114 does not necessarily have to be composed of multiple units according to the functions of the allocation server 112, and one control unit 114 may be configured to perform various functions.

[0027] (3) User terminal The user terminal 102 is a communication terminal that is used or managed by the user 100 and has communication and computing functions. Any known communication terminal can be used as the user terminal 102. The user terminal 102 may be, for example, a notebook or stationary computer, or may be a smartphone, a tablet computer, or a head-mounted portable communication terminal.

[0028] 3.Shipping method This delivery method will be described in detail below.

[0029] (1) Product ordering and delivery requests As shown in the scheme of FIG. 3A, first, the user 100 orders a product from the store 180 and requests delivery to a delivery box 190 that the user specifies in advance or when placing an order (hereinafter, the delivery box specified by the user 100 is referred to as a "specified delivery box"). At this time, the user 100 may request the store 180 to order and deliver the product verbally or by fax, or may request the store 180 to order and deliver the product using the user terminal 102. In the latter case, the user 100 uses the user terminal 102 to access a platform built by the store 180 or the platform provider 160, and requests the store 180 to order and deliver the product on the platform. The user 100 may also notify the store 180 of the desired date and time of delivery of the product directly or on the platform. When the store 180 confirms the order of the product and accepts the order, the order of the product is confirmed. The user 100 may specify a delivery box 190 other than the specified delivery box as an alternative delivery box in advance or when placing an order of the product. Alternatively, a physical space other than the designated delivery box may be specified as the delivery destination. The physical space may be a shopping mall (for example, a service counter in a shopping mall), a general retail store such as a supermarket or department store, a bookstore, a convenience store, a newspaper delivery station, or an amusement facility such as a pachinko parlor or a game center.

[0030] Thereafter, the store 180 uses the store server 182 to request the delivery company 170 to deliver the parcel including the product, directly or via the platform provider 160. That is, a delivery request signal is transmitted from the store server 182 to the delivery management server 172. The delivery request signal includes information about the parcel (parcel information). The parcel information includes the name, category, quantity, size, and management method of the product (for example, whether or not it needs to be refrigerated or frozen), and the delivery date and time desired by the user 100 (desired delivery date and time). The parcel information may include a box ID of the designated delivery box or an alternative delivery box. In addition, the parcel information may include information about the user 100 (address, name, telephone number, etc.), but when the parcel information includes a box ID of the designated delivery box or an alternative delivery box, the delivery destination is confirmed, so information about the user 100 does not need to be included in the parcel information. By not including information about the user 100 in the parcel information, leakage of the information can be prevented.

[0031] (2) Locker allocation and delivery The delivery company 170 that has been requested to perform delivery requests the allocation company 110 to allocate a locker. Specifically, the delivery company 170 uses the delivery management server 172 to access the allocation server 112 and transmits a locker allocation request signal to the allocation server 112. The locker allocation request signal includes package information and may also include a scheduled delivery date and time determined by the delivery company 170. If the package information does not include a box ID, the box ID of the delivery box 190 specified by the user 100 can be found on the delivery management server 172 from the address of the user 100, and the box ID can be included in the locker allocation request signal.

[0032] The allocation server 112 periodically or whenever a locker allocation request is received, accesses the designated delivery box or the box management server 152 to acquire locker information of the designated delivery box. Based on the acquired locker information and package information, the allocation business operator 110 selects a locker that is optimal for storing the package from among the available lockers. In selecting a locker, it is preferable to select a locker that matches the size of the package. For example, a locker is selected so that the volume of the package is 75% to 100%, 80% to 100%, 90% to 100%, or 95% to 100% of the locker's volume. In addition, when multiple packages are delivered to one user 100, a locker that can accommodate multiple packages may be selected. Alternatively, a locker may be selected according to the management method of the package. For example, in the case of a package that requires refrigeration, a locker equipped with a refrigeration function may be selected.

[0033] When a locker is selected, the allocation company 110 uses the allocation server 112 to transmit a locker allocation signal including the locker ID of the selected locker to the delivery management server 172 of the delivery company 170. This allows the delivery company 170 to secure a locker for the designated delivery box before delivering the package, so there is no shortage of lockers at the time of delivery and the package can be stored in the locker reliably. Therefore, the burden of re-delivering the package is not imposed.

[0034] After the package is delivered, a signal may be sent from the delivery management server 172 and / or the allocation server 112 to the user terminal 102 of the user 100 to notify the user terminal 102 that the package delivery has been completed. In addition, if the user 100 does not pick up the package even after a certain period of time (e.g., 24 hours, 2 days, 3 days, etc.) has passed since the package delivery has been completed, a urging notice to pick up the package may be sent from the delivery management server 172 and / or the allocation server 112 to the user terminal 102.

[0035] (3) Demand forecast As described above, the allocation company 110 selects an available locker from the designated delivery box in response to a locker allocation request from the delivery company 170, but there is not necessarily an available locker at that time. If there is no available locker, a locker cannot be allocated, so the delivery date and time cannot be confirmed and the user 100 cannot be notified of the delivery date and time. To avoid such a situation, in this delivery method, the allocation company 110 predicts demand for each delivery box 190 and adjusts delivery so that the total number of packages to be delivered does not exceed the capacity of the delivery box 190. Demand prediction may be performed periodically or each time a locker allocation request is received.

[0036] Demand forecasting is performed by the control unit 114 of the allocation server 112 or the demand forecasting unit 116 provided in the control unit 114, and the total number of parcels to be delivered per fixed period (e.g., daily, weekly, biweekly, monthly) is predicted. For example, demand forecasting can be performed using a supervised learning model. In demand forecasting using a supervised learning model, past usage records of multiple lockers provided in each delivery box 190 are used as training data. The past usage records are, for example, the total number of parcels actually stored in each delivery box 190 per fixed time period (e.g., one day, half a day) going back to the past fixed period (e.g., one year, six months, three months, etc.). The training data is collected for each delivery box 190 and stored in the memory unit 122.

[0037] In constructing a supervised learning model, various parameters can be used as input data (feature data). An example of a table summarizing the feature data is shown in FIG. 4. The feature data is linked to a box ID and a locker ID that identify a locker. Examples of the feature data include a store identifier (store ID) that identifies a store from which the product included in the package was ordered, an identifier (product ID) and category that identify the product, the size and number of packages, the time the package stayed in the locker, an identifier (delivery company ID) that identifies the delivery company that delivered the package, and the date and day of the week when the product was ordered. In addition, whether or not the user 100 used a benefit provided by the store 180 or the platform provider 160 when ordering the product may also be used as feature data. Although not shown in FIG. 4, the calendar (whether or not it is a public holiday, the day of the week on which the public holiday is, whether or not there is a consecutive holiday and the number of holidays during the consecutive holidays, etc.), whether or not there is a large-scale event (the Olympics, the World Cup, etc.), and the weather may also be used as feature data. The feature data is also collected for each delivery box 190 and stored in the storage unit 122.

[0038] The algorithm used in the supervised learning model can be selected arbitrarily, and for example, a machine learning algorithm can be used. Examples of machine learning algorithms include a linear regression model, a random forest, a decision tree, gradient boosting, and deep learning, and deep learning is particularly preferable. When deep learning is used, an algorithm such as a neural network, a deep neural network, or a convolutional neural network may be used as appropriate. By using a machine learning algorithm to predict demand for each delivery box 190, it is possible to predict the total number of packages to be stored in the delivery box 190 in a certain period of time. The difference between the predicted total number of packages and the training data, which is the past usage record, is the objective function, and the learning model can be trained and updated by minimizing this objective function.

[0039] (4) Promotion to users based on demand forecasts If the demand forecast indicates that the total number of packages will not exceed the capacity of the delivery box 190, the packages can be delivered according to the procedure described above. However, as shown in FIG. 5A, if a busy period is expected to arrive in which the total number of packages based on the demand forecast will exceed the capacity of the delivery box 190, the total number of packages may actually exceed the capacity of the delivery box 190 unless delivery adjustments are made, although this depends on the accuracy of the demand forecast (see the dashed line in FIG. 5A). If such a situation occurs, it will be impossible to assign lockers and it will be impossible to promptly notify the user 100 of the delivery date and time, leading to stagnation of delivery operations and a decrease in user satisfaction.

[0040] Therefore, in this delivery method, when a busy period is expected to arrive, the allocation company 110 conducts a promotion to adjust delivery to some or all of the users 100 who use the delivery box 190 (see FIG. 3A). Specifically, as shown in the flowchart in FIG. 6, when a product order is confirmed, the allocation server 112 determines whether the delivery date and time (i.e., the desired delivery date and time or the scheduled delivery date and time) falls within the busy period based on the demand forecast described above. If the delivery date and time does not fall within the busy period, a locker is assigned to each package according to the procedure described above, and the delivery company 170 delivers the package to the designated delivery box accordingly.

[0041] On the other hand, if the delivery date and time falls in a busy period, the allocation business 110 selects one or more users 100 who use the delivery box 190 and conducts a promotion. In other words, a request is made to change the delivery date and time and / or the delivery destination. The selection of the user 100 is performed by the control unit 114 of the allocation server 112 or the adjustment unit 118 provided in the control unit 114. The selected user 100 may be the user 100 who ordered the product, a user 100 who has already confirmed the order of the product but has not confirmed the delivery date and time, or a user 100 who has already confirmed the delivery date and time.

[0042] The selection of users 100 may be randomly (arbitrarily) selected from users 100 whose delivery dates fall within the busy period, but it is preferable to use user parameters to smoothly obtain the acceptance of the promotion by the users 100. The user parameters are stored in the memory unit 122 of the allocation server 112 as various parameters associated with the identifier (user ID) of each user 100. An example of parameters included in the user parameters is summarized in the table in FIG. 7. As shown in FIG. 7, the user parameters include the user's classification or attribute (e.g., whether the user is a single person or a general household), the past date and time of ordering a product and the date and day of delivery of the package, the time period during which the delivery was requested, the date and day of the week when the package was taken out of the locker, the time the package was left in the locker, whether or not the user requested door-to-door delivery in the past but was absent when the package was delivered (absentee record), the frequency of absence records, the type of residence (e.g., whether the user lives in a single-family home or an apartment building), and in the case of an apartment building, the floor on which the user lives, but is not limited to these. Furthermore, the user's behavior, specifically, the number of times promotions have been received from the allocation company 110 in the past and the frequency with which promotions have been accepted (number of times changes have been accepted / number of times changes have been requested), can also be used as user parameters.

[0043] The selection of users 100 may be performed, for example, by focusing on one user parameter (e.g., frequency of accepting promotions, floor of residence, etc.) and selecting users 100 according to that parameter. For example, the frequency of accepting promotions is an indicator of whether a user is cooperative or not. Also, if a user lives on a lower floor, it is more likely that the user will accept a promotion, since changing the delivery date and time or delivery address is not a big burden.

[0044] Alternatively, for more efficient selection, a supervised learning model may be used to select users 100 who are more likely to accept the promotion. That is, the above-mentioned multiple user parameters are used as features to predict the frequency of promotion acceptance using an algorithm such as a machine learning algorithm. A learning model is constructed and updated by minimizing an objective function that is the difference between the predicted result and the frequency of past promotion acceptance by users 100, which is the teaching data.

[0045] When a user 100 (hereinafter, referred to as a first user) to be promoted is selected, the allocation business operator 110 performs promotion on the first user (FIG. 6). That is, a change request signal for promotion is transmitted from the allocation server 112 to the user terminal 102 of the first user. When requesting a change in delivery date and time, for example, if the desired delivery date and / or scheduled delivery date is before the peak time of the total number of packages calculated by the demand forecast, a request is made to advance the delivery. Conversely, if the desired delivery date and / or scheduled delivery date and / or date is after the peak time of the total number of packages, a request is made to postpone the delivery. The same applies when the first user does not specify the desired delivery date and time, and the first user may be promoted to advance or postpone the delivery depending on the relationship between the scheduled delivery date and time and the peak time of the total number of packages. Alternatively, a request to advance or postpone the delivery may be made uniformly regardless of whether the desired delivery date and time or the scheduled delivery date and time is before or after the peak time of the total number of packages. Alternatively, a date and time when a locker suitable for the package is expected to be available based on the demand forecast may be set as the changed delivery date and time.

[0046] Alternatively, if there is another delivery box 190 where the total number of packages does not exceed the capacity during the busy period of the designated delivery box, the first user may be requested to change the delivery destination of the package to an alternative delivery box. Alternatively, if the first user has not designated an alternative delivery box, alternative delivery boxes may be selected in ascending order of distance from the designated delivery box designated by the first user, or the allocation server 112 may preferentially select alternative delivery boxes from delivery boxes 190 installed at stations or facilities close to the address of the first user.

[0047] (5) User response The first user who has received the promotion can accept or reject the promotion. Specifically, after receiving the change request signal, the first user uses the user terminal 102 to transmit a response signal to notify the allocation server 112 that the promotion is to be accepted or rejected.

[0048] When the first user accepts the promotion and the allocation server 112 receives a response signal notifying the promotion acceptance, a locker allocation signal including the changed delivery date and / or box ID of the changed delivery box 190 and the locker ID of the locker allocated by the allocation server 112 is generated by the allocation server 112 and transmitted to the delivery management server 172 of the delivery company 170. This reserves the delivery box 190 and its locker for package delivery. The delivery company 170 then delivers the package on the changed delivery date and / or to the changed delivery box 190.

[0049] If the first user accepts the promotion, the allocation company 110 may provide an incentive to the first user who accepted the promotion (FIG. 6). The incentive may be provided after or before the delivery of the package. The incentive may be provided by the control unit 114 of the allocation server 112 or the user management unit 120 provided in the control unit 114. The content of the incentive is arbitrary, and examples of the incentive include a discount on the fee for using the allocation company 110, a discount on the delivery fee, and an increase in points added when using the allocation company 110. Alternatively, a service of preferentially allocating a locker that is easier to use (for example, a locker located in a position that is easier to operate) may be provided as an incentive. When the incentive is granted, an incentive notification signal for notifying the first user of the grant may be transmitted from the allocation server 112 to the user terminal 102 of the first user.

[0050] On the other hand, if the first user rejects the promotion, another user 100 (hereinafter referred to as the second user) whose delivery date and time falls within the busy period may be selected and the promotion may be carried out (FIG. 6). The selection of the second user and the promotion method are the same as those for the first user. Furthermore, the delivery of the second user's parcel and the provision of incentives to the second user are also the same as those for the first user, so a description thereof will be omitted.

[0051] If the second user also rejects the promotion, the third and fourth users may be selected for the promotion. If both selected users reject the promotion, the delivery date and time of the parcel of the user 100 whose delivery date and time has not been confirmed may be set to a date after the locker of the delivery box 190 becomes available.

[0052] In this way, by predicting the demand for each delivery box 190 and adjusting the delivery so that the total number of delivered packages does not exceed the capacity of the delivery box 190, the number of delivered packages is leveled out as shown by the dashed line in Fig. 5B, which makes it possible to level out the workload of the delivery company 170. In addition, since a locker to store the package can be reliably reserved before delivery, it is possible to achieve a significant reduction in the cost and human resources required for redelivery.

[0053] Furthermore, by using user parameters in selecting users to whom a promotion is to be performed, it is possible to obtain smoother acceptance of the promotion. For example, promotions are preferentially selected for users 100 who tend to take out their packages promptly after they are placed in the delivery area (i.e., users 100 who do not spend much time with their packages) or users 100 who have frequently accepted promotions in the past. This reduces the frequency with which promotions are rejected and allows for quicker delivery adjustments, resulting in more efficient use of the delivery lockers.

[0054] In addition, when delivery adjustments are difficult or impossible, for example when a significant change in delivery date and time is required due to a long busy season, or when a suitable delivery box 190 or locker cannot be secured for an extended period of time, the above-mentioned physical space may be assigned as the delivery destination.

[0055] 4. Variations in delivery methods In the above example, the delivery of the package is requested to the delivery company 170. However, in this delivery method, as shown in Fig. 3B, the delivery request may be made directly to the allocation company 110. In this case, a delivery request signal for making the delivery request is sent from the store server 182 to the allocation server 112 via a platform established on the store server 182 or a platform established on the platform server 162. In addition, when a locker is secured, a delivery request for the package is sent from the allocation server 112 to the delivery management server 172 together with the locker ID of the allocated locker.

[0056] Alternatively, this delivery method can be applied even when the delivery entity is the store 180. Specifically, as shown in FIG. 3C, the store 180 requests the allocation company 110 to allocate a locker. That is, a locker allocation request signal is sent from the store server 182 to the allocation server 112 directly or via the platform of the platform server 162. Upon receiving the locker allocation request signal, the allocation server 112 selects a locker based on the locker information acquired from the delivery box 190 or the box management server 152, and sends the locker ID of the selected locker to the store server 182. The store 180 responsible for delivery simply stores the package in the allocated locker according to the locker ID.

[0057] 5. Signal transmission and reception The exchange of signals between the various components when delivering a parcel according to this delivery method will be described with reference to the flowcharts of Figures 8 to 10. In the following description, an example in which the user 100 orders a product on the store server 182 will be mainly described, but the product may be ordered in a face-to-face commercial transaction with the store 180 or on the platform server 162. When ordering a product on the platform server 162, the store server 182 can be replaced with the platform server 162.

[0058] (1) When delivery is handled by a delivery company FIG. 8 shows a flowchart for the case where the delivery company 170 is responsible for delivery, i.e., the delivery company 170 and the allocation company 110 are independent companies. First, the user 100 orders a product. Specifically, the user terminal 102 transmits a signal for ordering the product to the store server 182 (S100). Thereafter, the store server 182 may transmit a signal for confirming the order to the user terminal 102 (S102).

[0059] Thereafter, a package delivery request signal is transmitted from the store server 182 to the delivery management server 172 (S104). Although not shown, a signal for confirming receipt of the delivery request signal may be transmitted from the delivery management server 172 to the store server 182.

[0060] Thereafter, a locker allocation request signal is transmitted from the delivery management server 172 to the allocation server 112 (S106). Although not shown, a signal for confirming receipt of the locker allocation request signal may be transmitted from the allocation server 112 to the delivery management server 172.

[0061] As described above, each delivery box 190 periodically transmits locker information to the box management server 152 in response to a request from the box management server 152 or in response to the operation of a sensor provided in the delivery box 190 (S108). The allocation server 112 transmits a locker information request signal for transmitting locker information to the box management server 152 (S110). The locker information request signal includes the box ID of the designated delivery box and the alternative delivery box. In response to the locker information request signal, the box management server 152 transmits the locker information of the designated delivery box and the alternative delivery box to the allocation server 112 (S112). Note that, as described above, the locker information may be periodically transmitted from the box management server 152 to the allocation server 112. In this case, the allocation server 112 does not need to transmit the locker information request signal to the box management server 152.

[0062] Although not shown, if the desired delivery date and time or the scheduled delivery date and time of the package does not fall within a busy period based on the demand forecast by the allocation server 112, a locker allocation signal based on the acquired locker information is sent to the delivery management server 172. The delivery company 170 simply delivers the package to the locker corresponding to the locker ID included in the locker allocation signal.

[0063] On the other hand, if the desired delivery date and time or the scheduled delivery date and time of the parcel falls within the busy period, the allocation server 112 selects the user 100 to whom the promotion is to be applied. In the example shown in FIG. 8, the user 100 who ordered the product is selected. In this case, the allocation server 112 transmits a change request signal to the user terminal 102 to request a change to the delivery date and time or the delivery destination (S114). The user 100 who received the promotion transmits a signal (response signal) from the user terminal 102 to the allocation server 112 to notify that the promotion is accepted or rejected (S116). When the user 100 accepts the promotion, the allocation server 112 transmits a locker allocation signal including the changed delivery date and time and / or the box ID and locker ID of the alternative delivery box that is the changed delivery destination to the delivery management server 172 (S118). This ensures that a locker for delivering the parcel at the delivery date and time is secured, and the delivery company 170 can deliver the parcel according to the information included in the locker allocation signal. The delivery management server 172 may transmit a signal to the user terminal 102 to notify the user terminal 102 of the changed delivery date and / or delivery destination (S120). This signal may be transmitted from the allocation server 112 to the user terminal 102.

[0064] After the delivery is completed, the delivery management server 172 transmits a delivery notification signal to the user terminal 102 to notify that the delivery of the parcel has been completed (S122). Alternatively, the delivery notification signal may be transmitted from the delivery management server 172 to the allocation server 112, and then transmitted from the allocation server 112 to the user terminal 102. This allows the allocation server 112 to grasp the actual delivery date and time. When the parcel is delivered, locker information is transmitted from the delivery box 190 to the box management server 152 based on a signal detected by a sensor installed in the delivery box 190 (S124), and is further transmitted to the allocation server 112 (S126). Therefore, even by using the locker information, the allocation server 112 can grasp the date and time when the parcel was actually delivered and the date and time when the parcel was picked up by the user 100. The allocation server 112 stores these pieces of information in the storage unit 122 as user parameters, and can use them to select users for promotion.

[0065] Although not shown, when the sojourn time of the package exceeds a predetermined time, a reminder notice may be sent from the allocation server 112 to the user terminal 102. Also, an incentive notification signal may be sent from the allocation server 112 to the user terminal 102. The timing of sending the incentive notification signal is arbitrary, and it may be sent appropriately after receiving a response signal from the user terminal 102 of the user 100 who has accepted the promotion.

[0066] As described above, the user to whom the promotion is performed is not necessarily limited to the user 100 who ordered the product. For example, the promotion may be performed for a second user who, unlike the first user who ordered a certain product, has already confirmed the order of the product but has not yet confirmed the scheduled delivery date and time, or has already confirmed the scheduled delivery date and time. In addition, when a promotion is performed for a certain user (for example, the first user) but the promotion is rejected, the promotion may be performed for a second user different from the first user. In these cases, without transmitting a change request signal to the first user terminal 102-1 of the first user, or after receiving a response signal rejecting the promotion from the first user terminal 102-1 of the first user (S116), the allocation server 112 transmits a change request signal to the second user terminal 102-2 of the second user (S130) (see FIG. 9). The response signal is transmitted from the second user terminal 102-2 to the allocation server 112 (S132).

[0067] When a physical space other than the delivery box 190 is assigned as the delivery destination of the package, a delivery notification signal is transmitted to the user terminal 102 from a communication terminal owned by the entity that manages that physical space.

[0068] (2) When the allocation company is responsible for delivery As described above, in this delivery method, the allocation company 110 can be responsible for the delivery of packages. In other words, the delivery company 170 may be responsible for the allocation business. In this case, as shown in FIG. 10, a delivery request signal is sent from the store server 182 to the allocation server 112 (S140), and a receipt confirmation is sent to the store server 182 (S142). A delivery notification signal is also sent from the allocation server 112 to the user terminal 102 (S144). The exchange of signals other than these is the same as when the delivery company 170 and the allocation company 110 are independent companies, so a description thereof will be omitted.

[0069] As described above, in this delivery method, the allocation company 110, which receives a request from the delivery company 170, not only allocates a locker from the designated delivery box, but also selects one or more users 100 who use the designated delivery box based on the demand forecast of the delivery box 190, and performs a promotion for changing the delivery date and time or delivery destination for the selected users 100. In addition, since user parameters including the user's past behavior (such as the length of time the package is left in the delivery box, the frequency of absence, etc.) are used in the user selection, it is possible to provide the user with an incentive to accept the promotion. As a result, it is possible to make appropriate delivery adjustments when it is expected that the capacity of the delivery box 190 will be insufficient, and it is possible to avoid a situation in which redelivery is necessitated because all the lockers in the delivery box 190 are in use. This contributes to a significant reduction in the burden on the delivery person and the delivery cost.

[0070] The above-mentioned embodiments of the present invention may be combined as appropriate as long as they are not mutually inconsistent. In addition, the scope of the present invention also includes a display device in which the operator appropriately adds or removes components or modifies the design, or adds or omits a process or modifies conditions, based on the display device of each embodiment, as long as the gist of the present invention is maintained.

[0071] Even if there are other effects and advantages different from those brought about by the aspects of each of the above-mentioned embodiments, those which are clear from the description in this specification or which can be easily predicted by a person skilled in the art are naturally understood to be brought about by the present invention. [Explanation of symbols]

[0072] 100: user, 102: user terminal, 102-1: first user terminal, 102-2: second user terminal, 104: communication network, 110: allocation company, 112: allocation server, 114: control unit, 116: demand forecast unit, 118: adjustment unit, 120: user management unit, 122: storage unit, 124: communication unit, 126: input unit, 128: output unit, 150: box management company, 152: box management server, 160: platform provider, 162: platform server, 170: delivery company, 172: delivery management server, 180: store, 182: store server, 190: delivery box, 200: delivery system

Claims

1. A method for predicting the total number of packages to be delivered to a destination, the method comprising: a server connected to the destination in communication with the destination predicting the total number of packages to be delivered to the destination using a supervised learning model that uses usage history, which is the total number of packages delivered to the destination over a certain period of time in the past, as training data.

2. The method of claim 1, further comprising the server updating the supervised learning model by minimizing the difference between the predicted total number and the actual usage.

3. The method described in claim 1, wherein the delivery destination is a delivery box equipped with multiple lockers.

4. The method of claim 1, wherein the supervised learning model is constructed using at least one selected from a store identifier that identifies the store from which the product contained in the package was ordered, an identifier and category that identifies the product, the size and number of the package, the time that the package stayed at the delivery destination, an identifier that identifies the delivery company that delivered the package, the date and day of the week when the product was ordered, calendar, and weather as input data.

5. A server that is communicatively connected to a parcel delivery destination and is configured to predict the total number of parcels to be delivered to the destination using a supervised learning model that uses usage history, which is the total number of parcels delivered to the destination over a certain period of time in the past, as training data.

6. The server described in claim 5, further configured to update the supervised learning model by minimizing the difference between the predicted total number and the actual usage.

7. A server as described in claim 5, wherein the delivery destination is a delivery box equipped with multiple lockers.

8. The server described in claim 5, further configured to construct the supervised learning model using at least one selected from a store identifier identifying the store from which the product contained in the package was ordered, an identifier and category identifying the product, the size and number of the package, the time the package stayed at the delivery destination, an identifier identifying the delivery company that delivered the package, the date and day of the week when the product was ordered, calendar, and weather as input data.

Images