Information output method, information output device, and program

Abstract

This information output method includes: acquiring delivery destination information (current delivery information (112)) relating to at least one delivery destination to which an article is to be delivered by a moving body; acquiring product sales history information (115) relating to a product sales history for a user residing at the delivery destination; calculating purchasing probability information indicating a purchasing probability that the user will purchase merchandise, on the basis of the product sales history information (115); selecting a candidate article on the basis of the purchasing probability information; acquiring supplier information (merchandise information (201)) relating to a supplier of the candidate article; generating and then evaluating a delivery route candidate for the article to be delivered by the moving body and for the candidate article to be loaded onto the moving body, on the basis of the delivery destination information and the supplier information; and determining a delivery route candidate for which the evaluation result satisfies a prescribed condition, as a delivery route along which the moving body is to move.

Description

[Technical Field]

[0001] The present disclosure relates to an information output method for outputting a delivery route of a mobile object. [Background technology]

[0002] Patent document 1 discloses a method for generating a delivery plan for a delivery vehicle (mobile body) based on the parking location of the delivery vehicle and the delivery time after stopping, including the travel route between parking locations, the stopping time at each parking location, and the departure time from each parking location. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2019-149116 Summary of the Invention [Problem to be solved by the invention]

[0004] In recent years, there has been a study into utilizing the vacant space in delivery vehicles for merchandise sales. For this reason, delivery vehicles must also travel to suppliers to purchase merchandise for sale during deliveries.

[0005] Therefore, an object of the present disclosure is to provide an information output method and the like that can output the delivery route of a mobile object that sells goods while delivering them. [Means for solving the problem]

[0006] An information output method according to one aspect of the present disclosure is an information output method in an information output device, which acquires delivery destination information regarding at least one delivery destination to which goods are delivered by a mobile object, acquires product sales history information regarding the product sales history of users residing at the at least one delivery destination, calculates purchase probability information indicating the purchase probability of the user purchasing the goods based on the product sales history information, selects candidate goods that are the goods that the user may purchase based on the purchase probability information, acquires supplier information regarding suppliers of the candidate goods, generates candidate delivery routes for delivering the goods by the mobile object and loading the candidate goods onto the mobile object based on the delivery destination information and the supplier information, evaluates the candidate delivery routes based on at least one of the travel time, travel distance, and purchase probability information when the mobile object moves along the candidate delivery route, and determines and outputs the candidate delivery route whose evaluation result satisfies a predetermined condition as the delivery route along which the mobile object will travel.

[0007] and an output unit that outputs the delivery route. The information output device according to one aspect of the present disclosure includes: a delivery destination information acquisition unit that acquires delivery destination information regarding at least one delivery destination to which goods are delivered by a mobile object; a sales history information acquisition unit that acquires sales history information regarding the sales history of users residing at the at least one delivery destination; a purchase probability calculation unit that calculates purchase probability information indicating the purchase probability of the user purchasing the goods based on the sales history information; a selection unit that selects candidate goods that are the goods that the user may purchase based on the purchase probability information; a supplier information acquisition unit that acquires supplier information regarding suppliers of the candidate goods; a delivery route candidate generation unit that generates delivery route candidates for when the goods are delivered by the mobile object and when the candidate goods are loaded onto the mobile object based on the delivery destination information and the supplier information; an evaluation unit that evaluates the delivery route candidates based on at least one of the travel time, travel distance, and purchase probability information when the mobile object moves along the delivery route candidate, and determines the delivery route candidate whose evaluation result satisfies a predetermined condition as the delivery route along which the mobile object will travel; and an output unit that outputs the delivery route.

[0008] A program according to one aspect of the present disclosure is a program for causing a computer to execute the information output method.

[0009] These comprehensive or specific aspects may be realized as a system, an apparatus, a method, an integrated circuit, a computer program, or a non-transitory recording medium such as a computer-readable CD-ROM, or may be realized as any combination of a system, an apparatus, a method, an integrated circuit, a computer program, and a recording medium. [Effects of the Invention]

[0010] According to the present disclosure, it is possible to provide an information output method and the like that can output the delivery route of a mobile body that sells goods while delivering them. [Brief explanation of the drawings]

[0011] [Figure 1] FIG. 1 is a block diagram showing a configuration of an information output device according to an embodiment. [Figure 2] FIG. 2 is a table showing an example of current delivery information according to the embodiment. [Figure 3] FIG. 3 is a table illustrating an example of user information according to the embodiment. [Figure 4] FIG. 4 is a table illustrating an example of delivery history information according to the embodiment. [Figure 5] FIG. 5 is a table illustrating an example of sales history information according to the embodiment. [Figure 6] FIG. 6 is a flowchart showing the flow of the candidate item selection method according to the embodiment. [Figure 7] FIG. 7 is a flowchart showing the flow of the method for determining a delivery route according to the embodiment. [Figure 8] FIG. 8 is a schematic diagram showing a plurality of delivery route candidates according to the embodiment. [Figure 9] FIG. 9 is a block diagram showing a configuration of an information output device according to the second embodiment. [Figure 10]FIG. 10 is a block diagram showing a configuration of an information output device according to the third embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0012] (Findings that formed the basis of this disclosure) When deliveries are made by a delivery vehicle (mobile body), delivery items are removed from the delivery vehicle at the delivery destination, resulting in empty space in the delivery vehicle. To make effective use of this empty space, it has been considered to purchase merchandise for sale from suppliers during delivery, replenish the empty space, and recommend the merchandise to customers at the delivery destination. In this case, the delivery vehicle will travel from a delivery center to one or more delivery destinations and one or more suppliers before returning to the delivery center, so it is necessary to create an efficient delivery route. However, the above-mentioned Patent Document 1 does not take merchandise sales into consideration, and the reality is that it is not possible to create an efficient delivery route.

[0013] Therefore, the information output method according to the present disclosure is an information output method in an information output device, which acquires delivery destination information relating to at least one delivery destination to which goods are delivered by a mobile object, acquires product sales history information relating to the product sales history of users residing at the at least one delivery destination, calculates purchase probability information indicating the purchase probability of the user purchasing the goods based on the product sales history information, selects candidate goods that are the goods that the user may purchase based on the purchase probability information, acquires supplier information relating to suppliers of the candidate goods, generates candidate delivery routes for delivering the goods by the mobile object and loading the candidate goods onto the mobile object based on the delivery destination information and the supplier information, evaluates the candidate delivery routes based on at least one of the travel time, travel distance and purchase probability information when the mobile object moves along the candidate delivery route, and determines and outputs the candidate delivery route whose evaluation result satisfies a predetermined condition as the delivery route along which the mobile object will travel.

[0014] and an output unit that outputs the delivery route. The information output device according to the present disclosure further includes: a delivery destination information acquisition unit that acquires delivery destination information regarding at least one delivery destination to which goods are delivered by a mobile object; a sales history information acquisition unit that acquires sales history information regarding the sales history of users residing at the at least one delivery destination; a purchase probability calculation unit that calculates purchase probability information indicating the purchase probability of the user purchasing the goods based on the sales history information; a selection unit that selects candidate goods that are the goods that the user may purchase based on the purchase probability information; a supplier information acquisition unit that acquires supplier information regarding suppliers of the candidate goods; a delivery route candidate generation unit that generates delivery route candidates for when the goods are delivered by the mobile object and when the candidate goods are loaded onto the mobile object based on the delivery destination information and the supplier information; an evaluation unit that evaluates the delivery route candidates based on at least one of the travel time, travel distance, and purchase probability information when the mobile object moves along the delivery route candidate, and determines the delivery route candidate whose evaluation result satisfies a predetermined condition as the delivery route along which the mobile object will travel; and an output unit that outputs the delivery route.

[0015] Furthermore, a program according to the present disclosure is a program for causing a computer to execute the information output method.

[0016] According to this, candidate delivery routes are generated based on supplier information regarding suppliers of candidate items that the user of the delivery destination may purchase, and delivery destination information, so that candidate delivery routes can be created that take into consideration not only the delivery destination but also the supplier. Furthermore, candidate delivery routes are evaluated based on at least one of travel time, travel distance, and purchase probability information, and candidate delivery routes whose evaluation results satisfy predetermined conditions are determined and output as the delivery route, resulting in an efficient delivery route being output. Therefore, an efficient delivery route can be output for a mobile object that sells goods while delivering goods.

[0017] In addition, the purchase probability information may be calculated using a model trained based on the sales history information, delivery history information regarding delivery history for users residing at the at least one delivery destination, and user information regarding the users.

[0018] According to this method, purchase probability information is calculated using a model trained on sales history information, delivery history information related to delivery history for users residing at at least one delivery destination, and user information related to the users. Therefore, purchase probability information can be calculated using a model that takes into account sales history information, delivery history information, and user information. This increases the accuracy of purchase probability information, making it possible to select candidate items with a high purchase probability and obtain more appropriate supplier information. This increases the accuracy of delivery route candidate generation, resulting in more efficient delivery routes.

[0019] The model may also be retrained when certain update conditions are met.

[0020] According to this method, the model is re-learned when a predetermined update condition is met, so that the model can be updated to one that is appropriate for the current situation, thereby further improving the accuracy of the purchase probability information.

[0021] The information output method also groups users residing at a plurality of the delivery destinations, and selects the candidate items for each group based on the purchase probability information obtained for each group.

[0022] This allows users residing at multiple delivery destinations to be grouped, and candidate items to be selected for each group based on purchase probability information obtained for each group, thereby making it possible to select candidate items with a higher purchase probability.

[0023] Furthermore, in the information output method, when selecting the candidate item, the candidate item that is included in a predetermined top number of purchase probabilities included in the purchase probability information may be selected from among the plurality of candidate items.

[0024] According to this, from among a plurality of candidate items, candidate items that are included in the top predetermined number of purchase probabilities contained in the purchase probability information are selected, so that a plurality of candidate items that are more likely to be purchased can be selected.

[0025] In addition, the information output method may be such that, in evaluating the delivery route candidates, the higher the purchase probability included in the purchase probability information, the higher the evaluation.

[0026] According to this, the higher the purchase probability included in the purchase probability information, the higher the evaluation of the delivery route candidate, so that a delivery route candidate that makes it more likely that the candidate item will be purchased by the user is more likely to be determined as the delivery route.

[0027] In addition, the information output method may be such that, in evaluating the delivery route candidates, the shorter the travel time, the higher the evaluation.

[0028] According to this, the shorter the travel time, the higher the evaluation of the delivery route candidate is, so that a delivery route candidate with a short travel time is more likely to be determined as the delivery route.

[0029] In addition, the information output method may be such that, in evaluating the delivery route candidates, the shorter the travel distance, the higher the evaluation.

[0030] According to this, the shorter the travel distance, the higher the evaluation of the delivery route candidate is, so that a delivery route candidate with a short travel distance is more likely to be determined as the delivery route.

[0031] The candidate items may also include suggested items proposed by an item provider based on user information about the user.

[0032] According to this, since proposed products proposed by product providers are also included in the candidate products, product providers can appeal to users and sell candidate products.

[0033] The information output method may also output information encouraging the purchase of the candidate item loaded on the mobile object.

[0034] This allows information to be output that encourages the user to purchase the candidate item loaded on the mobile object, thereby increasing the user's desire to purchase.

[0035] In addition, the information output method may, when the delivery destination of one mobile body overlaps with the delivery destination of another mobile body, update the delivery routes of the one mobile body and the other mobile body so that the overlapping delivery destinations are consolidated into the one mobile body.

[0036] According to this, when a delivery destination of one mobile body overlaps with a delivery destination of another mobile body, the delivery routes of the one mobile body and the other mobile body are updated, and the overlapping delivery destinations are consolidated into one mobile body. Therefore, it is possible to have one mobile body make deliveries to the overlapping delivery destinations together, and the other mobile bodies not make deliveries. Therefore, it is possible to make the delivery routes of multiple mobile bodies more efficient.

[0037] In addition, the information output method may include, when the supplier of one mobile body overlaps with the supplier of another mobile body, consolidating the overlapping suppliers into the one mobile body, and updating the delivery routes of the one mobile body and the other mobile body so that the one mobile body and the other mobile body temporarily merge after the supplier.

[0038] According to this system, if a supplier of one mobile body overlaps with a supplier of another mobile body, the delivery routes of the one mobile body and the other mobile bodies are updated, the overlapping suppliers are consolidated into one mobile body, and the one mobile body and the other mobile bodies temporarily merge behind the supplier. As a result, one mobile body can purchase candidate items from the suppliers in bulk, and then hand over the candidate items to the other mobile body at the merge point. This makes it possible to improve the efficiency of the delivery routes of multiple mobile bodies.

[0039] The information output method may also include acquiring change information including a change in a situation on the delivery route, and updating the delivery route based on the change information.

[0040] According to this, when a change in the situation occurs on the delivery route, the delivery route is updated based on the change information, so that a delivery route that appropriately corresponds to the change in situation can be created.

[0041] The information output method may also include acquiring space information relating to a free space status of the mobile object, and updating the candidate items and the delivery route based on the space information.

[0042] This allows the candidate items and delivery routes to be updated according to the available space situation of the mobile object, so that candidate items and delivery routes suitable for the available space situation at that time can be created.

[0043] The information output method may also include acquiring margin information regarding margins for the candidate items of the plurality of moving bodies, and adjusting the spaces for the candidate items of each moving body based on the margin information of each moving body.

[0044] According to this, the space for candidate items of each mobile body is adjusted based on the margin information of the multiple mobile bodies, so that the loading capacity of candidate items can be adjusted according to the margin.

[0045] Furthermore, these comprehensive or specific aspects may be realized as a system, an apparatus, a method, an integrated circuit, a computer program, or a non-transitory recording medium such as a computer-readable CD-ROM, or may be realized as any combination of a system, an apparatus, a method, an integrated circuit, a computer program, and a recording medium.

[0046] (Embodiment) Hereinafter, the embodiments will be described in detail with reference to the drawings. Note that the embodiments described below are all comprehensive or specific examples. The numerical values, shapes, materials, components, component placement and connection forms, steps, and step order shown in the following embodiments are merely examples and are not intended to limit the scope of the claims. Furthermore, among the components in the following embodiments, components not described in the independent claims will be described as optional components.

[0047] [Embodiment 1] (Information output device) The information output device 100 according to the present disclosure is a device for managing deliveries of each of the autonomous vehicles, which are an example of a mobile object, when the autonomous vehicles deliver goods to the residences of multiple users. Furthermore, the information output device 100 is also a device for managing sales to each user in addition to deliveries. In other words, the information output device 100 has a function of planning and outputting efficient delivery routes for deliveries and sales made by each autonomous vehicle.

[0048] Fig. 1 is a block diagram showing the configuration of an information output device 100 according to an embodiment. As shown in Fig. 1, the information output device 100 is communicably connected to external systems, such as a merchandise inventory system 200 and an autonomous driving system 300, via the Internet.

[0049] Here, the product inventory system 200 is a system for managing multiple products (goods) that can be sold. The product inventory system 200 has product information 201 and inventory information 202. The product information 201 includes the address (location information) and contact information of the supplier of each product. Since each product also includes candidate products that the user can purchase, the product information 201 can be said to be an example of supplier information related to the suppliers of the candidate products. The inventory information 202 includes the inventory quantity of each product at each supplier.

[0050] The autonomous driving system 300 has road network information 301 within the delivery range of each autonomous driving vehicle. The road network information 301 includes various information necessary for autonomous driving, such as map information within the delivery range (including road maps, residential maps, etc.), traffic rule information, traffic regulation information, and congestion information.

[0051] The information output device 100 has a database unit 110 and a plan generation unit 120. The database unit 110 is a storage device such as an HDD or SSD, and has information used to plan delivery routes. Specifically, the database unit 110 has delivery vehicle information 111, current delivery information 112, user information 113, delivery history information 114, and product sales history information 115.

[0052] Delivery vehicle information 111 includes identification information (vehicle ID) of each autonomously driven vehicle used for delivery.

[0053] Current delivery information 112 is information that compiles together all deliveries for a given period (e.g., morning, afternoon, or one day) that have delivery destinations in roughly the same area. FIG. 2 is a table showing an example of current delivery information 112 according to an embodiment. As shown in FIG. 2, current delivery information 112 includes a user ID, which is user identification information, the user's address, the type of item to be delivered, the delivery date, and the desired delivery time zone. Current delivery information 112 is an example of delivery destination information related to the delivery destination.

[0054] The user information 113 includes information about each user residing at a plurality of delivery destinations. In other words, the user information 113 is an example of delivery destination information. FIG. 3 is a table showing an example of the user information 113 according to an embodiment. As shown in FIG. 3, the user information 113 includes a user ID, an address, a preference tendency, an age group, and a gender. The preference tendency is registered, for example, by the user answering a questionnaire. The user information 113 may also include other information that can be used when the user selects candidate items that are merchandise that the user may purchase.

[0055] The delivery history information 114 includes information on past deliveries to each user. Fig. 4 is a table showing an example of the delivery history information 114 according to the embodiment. As shown in Fig. 4, the delivery history information 114 includes a user ID, a vehicle ID, a date and time, an address, and a type of item for each delivery.

[0056] The sales history information 115 is information relating to the sales history of each user. FIG. 5 is a table showing an example of the sales history information 115 according to the embodiment. As shown in FIG. 5, the sales history information 115 includes a user ID, a vehicle ID, a date and time, an address, the product name of the product, the quantity of the product, and the type of the product. The sales history information 115 may also include other information that can be used when selecting candidate items.

[0057] 1, the database unit 110 acquires each piece of information input from an input unit (not shown). The input unit may be a keyboard, mouse, touch panel, etc. connected to the information output device 100, or a computer connected to the information output device 100 via the Internet. The database unit 110 classifies and stores each piece of acquired information into delivery vehicle information 111, current delivery information 112, user information 113, delivery history information 114, and product sales history information 115.

[0058] The plan generation unit 120 includes a CPU, RAM, and ROM, and the CPU loads a program stored in the ROM into the RAM and executes it. Specifically, the plan generation unit 120 includes a recommendation generation unit 121, a route planning unit 122, an evaluation unit 123, and an output unit 124.

[0059] The recommendation generation unit 121 is a part that generates candidate items that are merchandise that can be purchased by a user residing at a delivery destination. Specifically, the recommendation generation unit 121 has a first acquisition unit 1211, a purchase probability calculation unit 1212, and a selection unit 1213.

[0060] The first acquisition unit 1211 acquires the sales history information 115, the delivery history information 114, the user information 113, and the current delivery information 112 from the database unit 110, and also acquires the sales item information 201 and the inventory information 202 from the sales item inventory system 200. In other words, the first acquisition unit 1211 is an example of a sales history information acquisition unit that acquires sales history information.

[0061] The purchase probability calculation unit 1212 calculates purchase probability information indicating the purchase probability that a user will purchase a product, based on the product sales history information 115. Specifically, the purchase probability calculation unit 1212 creates a model trained on the basis of the product sales history information 115, the delivery history information 114, and the user information 113, and calculates the purchase probability using the model. Here, the model is an inference model such as a neural network model for performing inference processing. Furthermore, for example, the model may be a mathematical model for performing inference processing, a machine learning model, or a deep learning model.

[0062] Since such a model is created, it is possible to calculate a purchase probability for each merchandise item according to the merchandise sales history information 115, delivery history information 114, and user information 113. For example, since the merchandise sales history information 115 includes the product name, type, date and time, and quantity of merchandise items previously purchased by the user, it is possible to calculate a purchase probability that takes this purchasing tendency into consideration. Since the delivery history information 114 includes the type, date and time of past deliveries to the user, it is possible to calculate a purchase probability that takes this delivery tendency into consideration. Since the user information 113 includes the user's preference tendency, it is possible to calculate a purchase probability that takes this preference tendency into consideration.

[0063] The purchase probability calculation unit 1212 calculates purchase probability information for each product for each user by applying the current delivery information 112 and product information 201 to a model. Based on the calculation results, the purchase probability calculation unit 1212 creates a list for each user of products that are included in the top predetermined number of purchase probabilities included in the purchase probability information. Here, the top predetermined number of products refers to the products that are included in the first to Nth highest purchase probabilities.

[0064] The purchase probability calculation unit 1212 re-learns the model when predetermined update conditions are satisfied. The predetermined update conditions include a plurality of update conditions, and the purchase probability calculation unit 1212 re-learns the model when at least one of these update conditions is satisfied. Examples of the plurality of update conditions include when a predetermined amount of time has passed since the previous model was created, when the amount of change in purchase probability (described below) between the previous and current model is equal to or greater than a predetermined amount, when there is an increase or decrease in the types of merchandise included in the merchandise information 201, when there is an increase or decrease in the users included in the user information 113, etc.

[0065] The selection unit 1213 selects candidate items that are merchandise that the user may purchase based on the purchase probability information calculated by the purchase probability calculation unit 1212. Specifically, the selection unit 1213 first groups each user. The selection unit 1213 compares the lists of each user and classifies users whose lists contain merchandise that overlap with each other at a higher rate than a predetermined rate into the same group.

[0066] Next, the selection unit 1213 calculates the purchase probability of overlapping items for each group, and lists the items in descending order of purchase probability as a product candidate list. A product candidate list is created for each group, but the following explanation will only use the product candidate list for one group as an example. Similar processing is performed on the product candidate lists for other groups.

[0067] The selection unit 1213 extracts products that are currently in stock by comparing each product included in the product candidate list with the inventory information 202 acquired by the first acquisition unit 1211. The selection unit 1213 creates an in-stock candidate list by arranging the extracted products in descending order of purchase probability. The products included in this in-stock candidate list are candidate items for each user included in one group.

[0068] The route planning unit 122 is a component that plans delivery route candidates that include both delivery and sales. The route planning unit 122 has a second acquisition unit 1221 and a delivery route candidate generation unit 1222.

[0069] The second acquisition unit 1221 acquires current delivery information 112, which is an example of delivery destination information, from the database unit 110, and acquires merchandise information 201, which is an example of supplier information, from the merchandise inventory system 200. In other words, the second acquisition unit 1221 is an example of a delivery destination information acquisition unit that acquires delivery destination information and a supplier information acquisition unit that acquires supplier information. In addition, the second acquisition unit 1221 acquires road network information 301 from the autonomous driving system 300.

[0070] The delivery route candidate generation unit 1222 generates delivery route candidates for when an autonomous vehicle delivers goods and when candidate goods are loaded onto the autonomous vehicle, based on the current delivery information 112 and the merchandise information 201. Specifically, the delivery route candidate generation unit 1222 compares the in-stock candidate list with the merchandise information 201 to obtain the supplier address of each candidate item included in the in-stock candidate list. The delivery route candidate generation unit 1222 also obtains the address of the user who is the delivery destination from the current delivery information 112. The delivery route candidate generation unit 1222 links the supplier address of each candidate item and the address of the user who is the delivery destination to the road network information 301. Next, the delivery route candidate generation unit 1222 generates multiple delivery route candidates that can travel between delivery bases, each delivery destination, and each supplier on the road network information 301. At this time, the delivery route candidate generation unit 1222 may generate multiple delivery route candidates so as to cover all roads included in the road network information 301, or may generate multiple delivery route candidates so as to ensure that the distance between the routes is less than a predetermined distance. Note that a delivery base may also function as a supplier, and in this case, it can also be considered an example of a supplier.

[0071] The evaluation unit 123 evaluates the multiple delivery route candidates generated by the delivery route candidate generation unit 1222 and determines the one with the highest evaluation result as the delivery route that the autonomously driven vehicle will actually travel. Specifically, the evaluation unit 123 evaluates the delivery route candidates based on at least one of the travel time and travel distance when the autonomously driven vehicle travels along the delivery route candidate, and purchase probability information. For example, the evaluation unit 123 evaluates each delivery route candidate higher the higher the probability of purchasing candidate items that can be purchased from suppliers included in the delivery route candidate (first evaluation criterion). Furthermore, the evaluation unit 123 evaluates each delivery route candidate higher the shorter the travel time of the delivery route candidate (second evaluation criterion). In this case, the evaluation unit 123 may acquire (calculate) travel time taking into account traffic rule information, traffic regulation information, congestion information, etc. included in the road network information 301. Furthermore, the evaluation unit 123 evaluates each delivery route candidate higher the shorter the travel distance of the delivery route candidate (third evaluation criterion).

[0072] Note that evaluation criteria other than the first, second, and third evaluation criteria may also be used. Another evaluation criterion is an evaluation criterion that gives a higher evaluation to an autonomous vehicle as the free space of the autonomous vehicle decreases. When an autonomous vehicle only performs deliveries, the free space increases with each delivery from the delivery base. In other words, if a candidate delivery route stops at a supplier immediately after completing a delivery, the free space can be used to load candidate items, thereby reducing the free space. Furthermore, if a candidate item is sold at the delivery destination, the free space increases by the amount of the sale. In this case, if a candidate delivery route stops at a supplier immediately after a delivery destination where a user who is highly likely to purchase the candidate item loaded on the autonomous vehicle resides, the free space can be used to load a new candidate item, thereby reducing the free space. In this way, an evaluation criterion that gives a higher evaluation to an autonomous vehicle as the free space of the autonomous vehicle decreases is advantageous in that it can reduce free space.

[0073] The evaluation unit 123 may evaluate each delivery route candidate using at least one of these evaluation criteria, but a more appropriate evaluation can be performed when multiple evaluation criteria are used in combination. When multiple evaluation criteria are used, the weighting of the evaluation results of each evaluation criterion may be adjusted according to the importance of each evaluation criterion. The evaluation unit 123 determines the delivery route candidate that satisfies a predetermined condition from the evaluation results of each delivery route candidate as the delivery route that the autonomously driven vehicle will actually travel. In this embodiment, the highest evaluation result is exemplified as the predetermined condition, but other conditions may also be used.

[0074] The output unit 124 is a communication module that outputs the delivery route determined by the evaluation unit 123 to an external device via the Internet. For example, the output unit 124 outputs the delivery route to an autonomously driven vehicle via the Internet. As a result, the autonomously driven vehicle performs deliveries and sales based on the delivery route.

[0075] The output unit 124 may output to the autonomously driven vehicle information encouraging the user to purchase the candidate item, along with the delivery route determined by the evaluation unit 123. A notification unit (display unit or audio output unit) provided in the autonomously driven vehicle notifies the user of the information encouraging the user to purchase the candidate item, thereby increasing the user's desire to purchase the candidate item.

[0076] (Information output method) Next, we will explain the information output method in the information output device 100. The information output method is realized by the CPU of the plan generation unit 120 executing a program stored in the ROM. The information output method includes a method for selecting candidate items and a method for determining a delivery route.

[0077] First, a method for selecting candidate items will be described. Fig. 6 is a flowchart showing the flow of the method for selecting candidate items according to an embodiment. In this method for selecting candidate items, it is assumed that the first acquisition unit 1211 has already acquired sales history information 115, delivery history information 114, user information 113, current delivery information 112, sales item information 201, and inventory information 202, but the first acquisition unit 1211 may acquire each piece of information when it becomes necessary.

[0078] As shown in FIG. 6, in step S1, the purchase probability calculation unit 1212 determines whether or not a model exists. If a model has already been created, the process proceeds to step S2; if no model exists, the process proceeds to step S4.

[0079] In step S2, the purchase probability calculation unit 1212 determines whether or not a predetermined update condition is met, and if the predetermined update condition is met, proceeds to step S3, and if not, proceeds to step S5.

[0080] In step S3, the purchase probability calculation unit 1212 re-learns the existing model based on the sales history information 115, the delivery history information 114, and the user information 113.

[0081] In step S4, the purchase probability calculation unit 1212 creates a trained model based on the sales history information 115, the delivery history information 114, and the user information 113.

[0082] In step S5, the purchase probability calculation unit 1212 calculates purchase probability information for each product for each user by applying the current delivery information 112 and product information 201 to the model.

[0083] In step S6, the purchase probability calculation unit 1212 creates a list for each user that lists the products that are included in the top predetermined number of purchase probabilities included in the purchase probability information.

[0084] In step S7, the selection unit 1213 compares the lists L1 to Ln (n is the total number of users) of each user, and classifies users whose lists L1 to Ln include more than a predetermined percentage of overlapping items into the same group.

[0085] In step S8, the selection unit 1213 calculates the purchase probability of overlapping items for each group, and lists the items in descending order of purchase probability as item candidate lists C1 to Cm (m is the total number of groups).

[0086] In step S9, the selection unit 1213 extracts products that are currently in stock by comparing each product included in the product candidate lists C1 to Cm with the inventory information 202 acquired by the first acquisition unit 1211. The selection unit 1213 arranges the extracted products in order of highest purchase probability to create in-stock candidate lists Q1 to Qm. The products included in each of the in-stock candidate lists Q1 to Qm are candidate items for each user included in each group.

[0087] Next, a method for determining a delivery route will be described. Fig. 7 is a flowchart showing the flow of the method for determining a delivery route according to an embodiment. In this method for determining a delivery route, it is assumed that the second acquisition unit 1221 has already acquired the current delivery information 112 and merchandise information 201, but the second acquisition unit 1221 may acquire each piece of information when it becomes necessary.

[0088] As shown in FIG. 7, in step S11, the delivery route candidate generating unit 1222 sets i=1.

[0089] In step S12, the delivery route candidate generation unit 1222 compares the in-stock candidate list Qi of the group Gi with the sales product information 201 to obtain the supplier address of each candidate item included in the in-stock candidate list Qi.

[0090] In step S13, the delivery route candidate generating unit 1222 acquires the address of the user who is the delivery destination from the current delivery information 112.

[0091] In step S14, the delivery route candidate generation unit 1222 associates the address of the supplier of each candidate item and the address of the user who is the delivery destination with the road network information 301.

[0092] In step S15, the delivery route candidate generation unit 1222 generates a plurality of delivery route candidates that can travel between the delivery bases on the road network information 301, each delivery destination, and each supplier.

[0093] FIG. 8 is a schematic diagram showing multiple delivery route candidates according to an embodiment. In FIG. 8, all delivery route candidates RC1 to RC3 start from delivery base H, travel around each of delivery destinations Sh1 to Sh4 and each of suppliers Su1 and Su2, and return to delivery base H. Delivery route candidates RC1 to RC3 are represented by different line types. FIG. 8 illustrates an example in which each of delivery route candidates RC1 to RC3 connects each point (delivery base H, each of delivery destinations Sh1 to Sh4, each of suppliers Su1 and Su2) with a straight line, but multiple routes connecting each point are also generated according to the roads between those points. FIG. 8 illustrates an example in which there are three delivery route candidates, but in reality, multiple delivery route candidates are generated to cover all of the roads included in the road network information 301.

[0094] As shown in FIG. 7, in step S16, the evaluation unit 123 evaluates the multiple delivery route candidates generated by the delivery route candidate generation unit 1222, and selects the one with the highest evaluation result as the delivery route Ri based on the group Gi.

[0095] In step S17, the delivery route candidate generating unit 1222 adds 1 to i.

[0096] In step S18, the delivery route candidate generation unit 1222 determines whether i=m, and if i=m is not the case, proceeds to step S12, and if i=m, proceeds to step S19. As a result, delivery routes R1 to Rm are selected based on all groups G1 to Gm.

[0097] In step S19, the evaluation unit 123 determines the delivery route with the highest evaluation result from among the delivery routes R1 to Rm as the delivery route along which the autonomously driven vehicle will actually travel.

[0098] In step S20, the output unit 124 outputs the delivery route determined by the evaluation unit 123 to the autonomously driven vehicle via the Internet.

[0099] (Effects, etc.) As described above, the information output method according to the present disclosure acquires delivery destination information (current delivery information 112) relating to at least one delivery destination to which goods are delivered by a mobile body (autonomous driving vehicle), acquires merchandise history information 115 relating to the merchandise sales history of users residing at the at least one delivery destination, calculates purchase probability information indicating the purchase probability of the user purchasing the merchandise based on the merchandise sales history information 115, selects candidate goods that are merchandise that the user may purchase based on the purchase probability information, acquires supplier information (merchandise information 201) relating to suppliers of the candidate goods, generates candidate delivery routes for delivering the goods by the mobile body and loading the candidate goods onto the mobile body based on the delivery destination information and the supplier information, evaluates the candidate delivery routes based on at least one of the travel time, travel distance, and purchase probability information when the mobile body moves along the candidate delivery route, and determines and outputs the candidate delivery route whose evaluation result satisfies predetermined conditions as the delivery route along which the mobile body will travel.

[0100] The information output device 100 according to the present disclosure also includes a delivery destination information acquisition unit (second acquisition unit 1221) that acquires delivery destination information (current delivery information 112) related to at least one delivery destination to which an item is delivered by a mobile object (autonomous driving vehicle), a sales history information acquisition unit (first acquisition unit 1211) that acquires sales history information 115 related to the sales history of a user residing at at least one delivery destination, a purchase probability calculation unit 1212 that calculates purchase probability information indicating the probability that the user will purchase an item based on the sales history information 115, and a selection unit 1223 that selects candidate items that the user may purchase based on the purchase probability information. a supplier information acquisition unit (second acquisition unit 1221) that acquires supplier information related to the supplier of the candidate item; a delivery route candidate generation unit 1222 that generates delivery route candidates when the item is delivered by the mobile object and the candidate item is loaded onto the mobile object based on the delivery destination information and the supplier information; an evaluation unit 123 that evaluates the delivery route candidates based on the travel time, travel distance and purchase probability information when the mobile object moves along the delivery route candidate, and determines the delivery route candidate whose evaluation result satisfies predetermined conditions as the delivery route along which the mobile object will move; and an output unit 124 that outputs the delivery route.

[0101] Furthermore, a program according to the present disclosure is a program for causing a computer to execute the information output method.

[0102] According to this, candidate delivery routes are generated based on supplier information regarding suppliers of candidate items that the user of the delivery destination may purchase, and delivery destination information, so that candidate delivery routes can be created that take into consideration not only the delivery destination but also the supplier. Furthermore, candidate delivery routes are evaluated based on at least one of travel time, travel distance, and purchase probability information, and candidate delivery routes whose evaluation results satisfy predetermined conditions are determined and output as the delivery route, resulting in an efficient delivery route being output. Therefore, an efficient delivery route can be output for a mobile object that sells goods while delivering goods.

[0103] In addition, the purchase probability information is calculated using a model trained based on merchandise sales history information 115, delivery history information 114 relating to the delivery history of users residing at at least one delivery destination, and user information 113 relating to the users.

[0104] According to this, the purchase probability information is calculated using a model trained based on the sales history information 115, delivery history information 114, and user information 113, so that the purchase probability information can be calculated using a model that takes into account the sales history information 115, delivery history information 114, and user information 113. This increases the accuracy of the purchase probability information, making it possible to select candidate items with a high purchase probability and obtain more appropriate supplier information. This increases the accuracy of the generation of delivery route candidates, resulting in more efficient delivery routes.

[0105] The model is also retrained when certain update conditions are met.

[0106] According to this method, the model is re-learned when a predetermined update condition is met, so that the model can be updated to one that is appropriate for the current situation, thereby further improving the accuracy of the purchase probability information.

[0107] Furthermore, users residing at multiple delivery destinations are grouped, and candidate items are selected for each group based on purchase probability information obtained for each group.

[0108] This allows users residing at multiple delivery destinations to be grouped, and candidate items to be selected for each group based on purchase probability information obtained for each group, thereby making it possible to select candidate items with a higher purchase probability.

[0109] Furthermore, when selecting a candidate item, the candidate item that is included in the top predetermined number of purchase probabilities included in the purchase probability information is selected from among a plurality of candidate items.

[0110] According to this, from among a plurality of candidate items, candidate items that are included in the top predetermined number of purchase probabilities contained in the purchase probability information are selected, so that a plurality of candidate items that are more likely to be purchased can be selected.

[0111] In addition, in the evaluation of the delivery route candidates, the higher the purchase probability included in the purchase probability information, the higher the evaluation.

[0112] According to this, the higher the purchase probability included in the purchase probability information, the higher the evaluation of the delivery route candidate, so that a delivery route candidate that makes it more likely that the candidate item will be purchased by the user is more likely to be determined as the delivery route.

[0113] In addition, when evaluating delivery route candidates, the shorter the travel time, the higher the evaluation.

[0114] According to this, the shorter the travel time, the higher the evaluation of the delivery route candidate is, so that a delivery route candidate with a short travel time is more likely to be determined as the delivery route.

[0115] In addition, when evaluating delivery route candidates, the shorter the travel distance, the higher the evaluation.

[0116] According to this, the shorter the travel distance, the higher the evaluation of the delivery route candidate is, so that a delivery route candidate with a short travel distance is more likely to be determined as the delivery route.

[0117] Also, information encouraging the purchase of candidate items loaded on the vehicle is output.

[0118] This allows information to be output that encourages the user to purchase the candidate item loaded on the mobile object, thereby increasing the user's desire to purchase.

[0119] [Embodiment 2] Next, a second embodiment will be described. In the following description, parts equivalent to those in the first embodiment will be given the same reference numerals, and the description thereof may be omitted. In this second embodiment, a case will be described in which a suggested item proposed by an item provider is included in the candidate items. FIG. 9 is a block diagram showing the configuration of an information output device 100 according to the second embodiment.

[0120] Here, the term "goods provider" refers to a trader that handles various goods, and includes, for example, a supplier and a wholesaler that delivers goods to the supplier. As shown in Fig. 9, information terminals 500 owned by a plurality of goods providers are connected to the information output device 100 according to the second embodiment so as to be able to communicate freely via the Internet. The information terminals 500 include, for example, PCs, tablet terminals, smartphones, and feature phones.

[0121] The information output device 100 provides user information of each user included in the determined delivery route to an information terminal 500 of a supplier located on or near the determined delivery route, or of a wholesaler that does business with the supplier. The information terminal device 500 selects suggested products that the user can purchase based on various information included in the user information, and transmits suggested information (product information, inventory information, supplier address, etc.) related to the suggested products to the information output device 100. The information output device 100 also manages the suggested products as candidate products based on the suggested information. This makes it possible for suggested products proposed by product providers to be sold in autonomous vehicles. In other words, product providers can sell products (new products) that they want to promote to users in autonomous vehicles.

[0122] In addition, in an autonomously driven vehicle, the suggested item and other candidate items may be loaded in different spaces.

[0123] In addition, the information output method may not adopt the suggested item if a candidate item with a high purchase probability for the user at the delivery destination is available, and may adopt the suggested item if a candidate item with a high purchase probability for the user at the delivery destination is not available.

[0124] [Embodiment 3] Next, a third embodiment will be described. In the following description, parts equivalent to those in the first embodiment will be assigned the same reference numerals, and their description may be omitted. For example, as delivery progresses, the amount of available space in the autonomous vehicle tends to increase. The more available space there is, the more candidate items can be loaded onto the autonomous vehicle. In other words, it becomes possible to propose many candidate items to users later on the delivery route. In this third embodiment, a case will be described in which the candidate items and delivery route are updated taking into account the available space situation in the autonomous vehicle. FIG. 10 is a block diagram showing the configuration of an information output device 100 according to the third embodiment.

[0125] As shown in FIG. 10, a space information acquisition unit 600 provided in the autonomously driven vehicle is communicatively connected to the information output device 100 via the Internet. The space information acquisition unit 600 is a device that acquires space information for the autonomously driven vehicle. The space information is information including the available space status for the autonomously driven vehicle. The space information acquisition unit 600 includes a device equipped with a camera that captures images of the loading space of the autonomously driven vehicle and an analysis unit that automatically acquires the available space status by analyzing the image obtained from the camera. Note that the space information acquisition unit 600 may also be equipped with a camera that captures images of the loading space of the autonomously driven vehicle, and the image captured by the camera may be used as the space information. In this case, in the information output device 100A, an operator may determine the available space status by visually checking the image that is the space information.

[0126] The recommendation generation unit 121 of the information output device 100 successively acquires space information and updates the candidate items for users who reside at the delivery destination from that point onward based on this space information. At this time, the candidate items may be updated taking into consideration only the users at the delivery destination from that point onward. The candidate items may also be updated taking into consideration the purchase probability, the user's allowed arrival time, etc.

[0127] Similarly, the route planning unit 122 also successively acquires space information, and plans delivery route candidates from that point onwards based on this space information and updated candidate items. The evaluation unit 123 evaluates the multiple delivery route candidates generated by the delivery route candidate generation unit 1222, and determines the one with the highest evaluation result as the delivery route that the autonomously driven vehicle will actually travel.

[0128] Note that contact information (such as telephone number and email address) of each user may be registered in the information output device 100. When a candidate item is loaded in an empty space, the information output device 100 may notify the contact information of the user to whom the delivery is about to be made of the candidate item that the candidate item has been loaded. This allows the user to consider purchasing the candidate item in advance.

[0129] In addition, since multiple deliveries may be delivered to the same delivery destination, the loading capacity of the candidate items may be determined based on the amount of free space remaining after delivery.Since the size of the loading space of the autonomous vehicle may vary, the candidate items may be determined based on the amount of free space remaining after delivery.

[0130] For example, if a user with a high purchase probability does not purchase a specific candidate item, and subsequent users have a low purchase probability, it can be inferred that the specific candidate item will remain unsold. When this inference is made, the candidate item and delivery route may be updated so that the candidate item is replaced with a new candidate item with a high purchase probability for other users. The replacement location is a supplier that handles the new candidate item. When replacing a candidate item, the location where the candidate item is unloaded from the autonomous vehicle may be different from the location where the new candidate item is purchased. The location where the candidate item is unloaded may be the location where the candidate item was purchased, or may be a different location (e.g., a different supplier of the same company).

[0131] When the autonomous vehicle has completed all deliveries, it returns to delivery center H. However, if unsold candidate items remain on the autonomous vehicle, they will interfere with the loading of the next delivery. For this reason, the delivery route may be determined taking into consideration unloading unsold candidate items from the autonomous vehicle. For example, a location where the candidate items can be unloaded may be determined on the route from the last delivery destination on the delivery route back to delivery center H. If no location where the candidate items can be unloaded is found on the route from the last delivery destination on the delivery route back to delivery center H, a location where the candidate items can be unloaded may be determined on the route from the second-to-last delivery destination to the last delivery destination, or on the route from the third-to-last delivery destination to the second-to-last delivery destination. In other words, a location where the candidate items can be unloaded may be determined on the final route of the delivery route. The final route of the delivery route refers to the route from the nth (2≦n≦m / 2: n is an integer) delivery destination from the last to the n-1th delivery destination from the last, assuming that there are m delivery destinations in total on the delivery route. The location at which the candidate item is to be dropped off may be determined based on the probability that the user at the destination will purchase the candidate item. For example, if it is determined that the probability of the candidate item being purchased at both the last and second-to-last destinations is low, it is determined that the candidate item will be dropped off at the third or subsequent destination on the delivery route.

[0132] There is also a possibility that the candidate item will be sold out at the final delivery destination. In anticipation of this case, the information output method may determine, based on the purchase probability of the candidate item at the final delivery destination, whether to drop off the candidate item on the route from the second-to-last delivery destination to the final delivery destination, or to deliver the candidate item to the final delivery destination without dropping off the candidate item. Note that if the candidate item remains unsold at the final delivery destination despite having been delivered to the final delivery destination, a location outside the delivery route where the candidate item can be dropped off may be determined. In this case, whether or not to adopt that location may be determined depending on the travel cost to that location. If the candidate item can be dropped off at delivery base H, such a determination process is unnecessary.

[0133] In addition, as in the second embodiment, when the information terminals 500 owned by multiple product providers are connected to the information output device 100 via the Internet and can communicate freely, the information terminals 500 may be notified based on the space information, and product suggestions may be solicited. For example, if an increase in free space occurs in an autonomous vehicle during delivery, there is a desire to make effective use of that free space. In other words, when the space information indicates that free space has increased, the information output device 100 notifies the information terminals 500 of suppliers on the subsequent delivery route and solicits candidate products. This allows the supplier that has applied to load the candidate products onto the autonomous vehicle.

[0134] The product provider may register suggested products in advance. In this case, when it is found from the space information that available space has increased, the information output device 100 may include a supplier of the product provider that handles the registered suggested products in the delivery route if the supplier is on the delivery route. Note that the product provider may register suppliers in advance. In this case, the suggested products are determined after the supplier is included in the delivery route.

[0135] In this embodiment, an autonomous vehicle that delivers goods has been described as an example of a mobile body. However, an autonomous vehicle for mobile sales (hereinafter referred to as a mobile sales vehicle) may also be used as the mobile body. In this case, when it is determined that there is an increase in available space based on the space information of the mobile sales vehicle, the information output device 100 may deliver goods to a user who requests delivery on the route to the next destination (next sales destination, base). In this case, if the delivery destination and delivery source are on the same route (neighborhood on the route), the delivery request for goods may be accepted.

[0136] Since it is not possible to know in advance how much free space will be available during transport, the amount of free space can be determined based on the sales situation at the destination, and the conditions for deliverable items (e.g., type, size, quantity, etc.) can be determined based on the results of that determination.

[0137] It is also possible that a mobile sales vehicle may stop at a warehouse during its travels to replenish merchandise. This replenishment may be taken into consideration when deciding whether to accept a delivery request for goods. Warehouses include not only fixed facilities but also mobile warehouses (such as trucks). The location where the mobile warehouse and the mobile sales vehicle meet may be considered as a replenishment location when planning a delivery route.

[0138] [Embodiment 4] Next, a fourth embodiment will be described. In the following description, parts equivalent to those in the first embodiment will be given the same reference numerals, and their description may be omitted. The first embodiment described above illustrates a case where a delivery route for one autonomously driven vehicle is determined. In this fourth embodiment, a case where delivery routes for multiple autonomously driven vehicles are determined will be described.

[0139] First, the information output device 100 collects delivery requests to be delivered within a predetermined period (for example, one day) and determines which delivery to allocate to which autonomous vehicle. As a result, current delivery information 112 for each autonomous vehicle is created. Based on the current delivery information 112 for each autonomous vehicle, the information output device 100 creates and acquires margin information regarding the margin for candidate items for each autonomous vehicle. The margin for candidate items is an index that indicates how much space can be secured for each autonomous vehicle for the candidate items. The margin for an autonomous vehicle can also be considered an index that indicates that another delivery must be made immediately after the current delivery.

[0140] Since the main role of autonomous vehicles is delivery, it is first necessary to secure space for delivery. In other words, the space for candidate items can be considered as excess space. For example, suppose an autonomous vehicle has a total of three item installation spaces. If there is a high degree of margin, there will be two spaces for candidate items; if there is a medium degree of margin, there will be one space for candidate items; and if there is a low degree of margin, there will be no space for candidate items.

[0141] The information output device 100 determines a delivery route for each autonomous vehicle based on the margin information of each autonomous vehicle. At this time, the autonomous vehicle with a higher margin has more space for candidate items, and therefore a delivery route with a larger load capacity for candidate items is determined. The autonomous vehicle with a low margin has zero space for candidate items, and therefore a delivery route that does not purchase candidate items is determined.

[0142] Furthermore, the number of times a candidate item is purchased (number of times it visits a supplier) may be limited depending on the autonomous vehicle's margin of safety. For example, if an autonomous vehicle has a total of three item installation spaces, it will have at least two opportunities to purchase the candidate item since it will visit three delivery destinations. In this case, an autonomous vehicle with a high margin of safety can be limited to two purchases, while an autonomous vehicle with a medium margin of safety can be limited to one purchase.

[0143] Here, since delivery demand fluctuates depending on the time of day, etc., if the information output device 100 can predict the margin of each autonomous vehicle in advance, the information output device 100 may adjust the space for candidate items based on the predicted margin. Since the margin of each autonomous vehicle changes from moment to moment, the space for candidate items may be adjusted in accordance with the changed margin.

[0144] Furthermore, when the autonomous vehicle has a high margin of error, delivery efficiency will not decrease even if the autonomous vehicle takes a detour to purchase candidate items, so the delivery route may be determined to cover a wider range of available supplies than when the margin of error is low.

[0145] Additionally, the space for candidate items may be adjusted depending on the level of delivery busyness, which may be determined, for example, by the number of delivery requests and the number of autonomous vehicles available to handle them.

[0146] (others) Although the information output method according to the embodiment of the present invention has been described above, the present invention is not limited to the above embodiment. In other words, the embodiment disclosed herein is illustrative in all respects and is not restrictive. The scope of the present invention is defined by the claims, and it is intended to include all modifications within the meaning and scope of the claims.

[0147] For example, the information output method may update the delivery routes of one mobile body and the other mobile body when a delivery destination of one mobile body overlaps with a delivery destination of another mobile body so that the overlapping delivery destinations are consolidated into one mobile body. Specifically, after a delivery route for one mobile body and a delivery route for the other mobile body are determined, an evaluation unit compares these delivery routes and determines whether the delivery destinations overlap. If an overlap exists, a delivery route candidate generation unit creates multiple delivery route candidates for the one mobile body and multiple delivery route candidates for the other mobile body so that the overlapping delivery destinations are consolidated into one mobile body. The evaluation unit then evaluates each delivery route candidate to determine a delivery route for the one mobile body and a delivery route for the other mobile body.

[0148] According to this, when a delivery destination of one mobile body overlaps with a delivery destination of another mobile body, the delivery routes of the one mobile body and the other mobile body are updated, and the overlapping delivery destinations are consolidated into one mobile body. Therefore, it is possible to have one mobile body make deliveries to the overlapping delivery destinations together, and the other mobile bodies not make deliveries. Therefore, it is possible to make the delivery routes of multiple mobile bodies more efficient.

[0149] Furthermore, the information output method may update the delivery routes of one mobile body and the other mobile body so that, when a supplier of one mobile body overlaps with a supplier of another mobile body, the overlapping suppliers are consolidated into one mobile body and the one mobile body and the other mobile body temporarily join behind the supplier. Specifically, after a delivery route for one mobile body and a delivery route for the other mobile body are determined, an evaluation unit compares these delivery routes to determine whether the suppliers overlap. If an overlap exists, a delivery route candidate generation unit generates multiple delivery route candidates for the one mobile body and multiple delivery route candidates for the other mobile body so that the overlapping suppliers are consolidated into one mobile body and the one mobile body and the other mobile body temporarily join behind the supplier. The evaluation unit then evaluates each delivery route candidate to determine a delivery route for the one mobile body and a delivery route for the other mobile body.

[0150] According to this system, if a supplier of one mobile body overlaps with a supplier of another mobile body, the delivery routes of the one mobile body and the other mobile bodies are updated, the overlapping suppliers are consolidated into one mobile body, and the one mobile body and the other mobile bodies temporarily merge behind the supplier. As a result, one mobile body can purchase candidate items from the suppliers in bulk, and then hand over the candidate items to the other mobile body at the merge point. This makes it possible to improve the efficiency of the delivery routes of multiple mobile bodies.

[0151] The information output method may also acquire change information including status changes along the delivery route and update the delivery route based on the change information. Specifically, when the second acquisition unit acquires the change information, the delivery route candidate generation unit creates multiple delivery route candidates based on the status change information. The evaluation unit then evaluates each delivery route candidate to determine the delivery route for the mobile object. Here, the change information includes information on an increase or decrease in the number of delivery destinations due to cancellations or additions of deliveries, inventory changes for candidate items at suppliers, changes in traffic regulation information or congestion information included in the road network information 301, information on the absence of users at delivery destinations, and information on changes in the desired delivery date and time.

[0152] According to this, when a change in the situation occurs on the delivery route, the delivery route is updated based on the change information, so that a delivery route that appropriately corresponds to the change in situation can be created.

[0153] In the above embodiment, a case has been exemplified in which candidate items that a user may purchase are selected based on the purchase probability. However, candidate items may be selected taking into consideration factors other than the purchase probability, such as the price of the merchandise, the priority of the user, the priority of the merchandise, and users with early delivery times. For example, in the case of the price of the merchandise, priority is given to merchandise with a higher price. In the case of the priority of the user, priority is given to users with a higher contract fee. In the case of the priority of the merchandise, priority is given to suppliers with a higher contract fee.

[0154] Furthermore, in the information output method executed by the information output device, when a delivery destination of one mobile body overlaps with a delivery destination of another mobile body, the candidate items to be sold at the overlapping delivery destination may be different between the one mobile body and the other mobile body, thereby preventing the same candidate items from being recommended to the overlapping delivery destinations.

[0155] Furthermore, the information output method executed by the information output device may output information that provides more installation space for candidate items in the autonomous driving vehicle to users who have an extremely high probability of purchase (for example, 90% or more).

[0156] Furthermore, in the above embodiment, an autonomous vehicle is exemplified as a moving body, but the moving body may also be an automobile, motorcycle, bicycle, etc. driven by a driver. In this case, the output unit may output the delivery route to a display terminal provided on the automobile, motorcycle, bicycle, etc., or to a mobile terminal held by the driver.

[0157] Although aspects of the information output device have been described above based on the embodiments, the aspects of the information output device are not limited to the embodiments. Modifications conceivable by those skilled in the art may be applied to the embodiments, and multiple components in the embodiments may be combined in any manner. For example, a process performed by a specific component in the embodiments may be performed by another component instead of the specific component. Furthermore, the order of multiple processes may be changed, or multiple processes may be performed in parallel.

[0158] Furthermore, the information output method including the steps performed by each component of the information output device may be executed by any device or system. That is, the information output method may be executed by the information output device or by another device or system.

[0159] For example, the above-described information output method may be executed by a computer including a processor, a memory, an input / output circuit, etc. In this case, the information output method may be executed by the computer executing a program for causing the computer to execute the information output method. Furthermore, the program may be recorded on a non-transitory computer-readable recording medium.

[0160] Furthermore, the multiple components of the information output device may be configured with dedicated hardware, general-purpose hardware that executes the above-mentioned programs, or a combination of these. The general-purpose hardware may be configured with a memory that stores the programs and a general-purpose processor that reads and executes the programs from the memory. Here, the memory may be a semiconductor memory or a hard disk, and the general-purpose processor may be a CPU.

[0161] Furthermore, the dedicated hardware may be configured with a memory and a dedicated processor, etc. For example, the dedicated processor may refer to the memory and execute the information processing method described above.

[0162] Furthermore, each component of the information output device may be an electric circuit. These electric circuits may form a single electric circuit as a whole, or each may be a separate electric circuit. Furthermore, these electric circuits may correspond to dedicated hardware, or may correspond to general-purpose hardware that executes the above-mentioned programs, etc. [Industrial Applicability]

[0163] The present disclosure is applicable to an information output method for creating and outputting a delivery route for a mobile object. [Explanation of symbols]

[0164] 100 Information output device 110 Database Department 111 Delivery vehicle information 112 Current delivery information (delivery destination information) 113 User Information 114 Delivery history information 115 Merchandise sales history information 120 Plan Generation Unit 121 Recommendation Generation Unit 122 Route Planning Department 123 Evaluation Department 124 Output section 200 Merchandise Inventory System 201 Product Information (Supplier Information) 202 Stock Information 300 Autonomous Driving System 301 Road Network Information 500 Information terminal 600 Space Information Acquisition Department 1211 First Acquisition Department (Merchandise Sales History Information Acquisition Department) 1212 Purchase Probability Calculation Unit 1213 Selection Department 1221 Second acquisition unit (delivery destination information acquisition unit, supplier information acquisition unit) 1222 Delivery route candidate generation unit H Delivery Center Sh1~Sh4 Delivery address Su1, Su2 Suppliers

Claims

1. An information output method in an information output device, acquiring delivery destination information regarding at least one delivery destination to which the item is delivered by the mobile object; Acquire product sales history information regarding product sales histories for users residing at the at least one delivery destination; calculating purchase probability information including a purchase probability that the user will purchase each of the merchandise items based on the merchandise sales history information; Selecting candidate items that are the merchandise that the user may purchase based on the purchase probability information; acquiring supplier information regarding a supplier of the candidate item; generating delivery route candidates for when the item is delivered by the mobile body and when the candidate item is loaded onto the mobile body based on the delivery destination information and the supplier information; The delivery route candidates are evaluated based on at least one of the travel time, travel distance, and purchase probability information when the mobile object travels along the delivery route candidates, and the delivery route candidate whose evaluation result satisfies a predetermined condition is determined as the delivery route along which the mobile object will travel, and output. Information output method.

2. The purchase probability information is calculated using a model trained based on the product sales history information, delivery history information relating to delivery histories of users residing at the at least one delivery destination, and user information relating to the users. The information output method according to claim 1 .

3. The model is retrained when a predetermined update condition is met. The information output method according to claim 2 .

4. Grouping users residing at a plurality of the delivery destinations, and selecting the candidate items for each group based on the purchase probability information obtained for each group. The information output method according to any one of claims 1 to 3.

5. When selecting the candidate product, the candidate product that is included in a predetermined top number of purchase probabilities included in the purchase probability information is selected from the plurality of candidate products. The information output method according to any one of claims 1 to 3.

6. In the evaluation of the delivery route candidates, the higher the purchase probability included in the purchase probability information, the higher the evaluation. The information output method according to any one of claims 1 to 3.

7. The candidate products also include suggested products proposed by product providers based on user information about the user. The information output method according to any one of claims 1 to 3.

8. In the evaluation of the delivery route candidates, the shorter the travel time, the higher the evaluation. The information output method according to any one of claims 1 to 3.

9. In the evaluation of the delivery route candidates, the shorter the travel distance, the higher the evaluation. The information output method according to any one of claims 1 to 3.

10. Outputting information encouraging the purchase of the candidate item loaded on the mobile body The information output method according to any one of claims 1 to 3.

11. If the delivery destination of one mobile body overlaps with the delivery destination of another mobile body, the delivery routes of the one mobile body and the other mobile body are updated so that the overlapping delivery destinations are consolidated into the one mobile body. The information output method according to any one of claims 1 to 3.

12. If the supplier of one mobile body overlaps with the supplier of another mobile body, the overlapping suppliers are consolidated into the one mobile body, and the delivery routes of the one mobile body and the other mobile body are updated so that the one mobile body and the other mobile body temporarily join after the supplier. The information output method according to any one of claims 1 to 3.

13. Acquire change information including a change in the situation on the delivery route; Update the delivery route based on the change information The information output method according to any one of claims 1 to 3.

14. Obtaining space information regarding the available space status of the mobile body; Update the candidate items and the delivery routes based on the space information The information output method according to any one of claims 1 to 3.

15. acquire margin information regarding margins of the plurality of moving bodies with respect to the candidate items; Adjusting the space for the candidate item on each moving body based on the margin information on each moving body. The information output method according to any one of claims 1 to 3.

16. a delivery destination information acquisition unit that acquires delivery destination information regarding at least one delivery destination to which an item is delivered by a mobile object; a sales history information acquisition unit that acquires sales history information regarding sales histories of users residing at the at least one delivery destination; a purchase probability calculation unit that calculates purchase probability information including a purchase probability of each product being purchased by the user based on the product sales history information; a selection unit that selects candidate items that are the merchandise that the user may purchase based on the purchase probability information; a supplier information acquisition unit that acquires supplier information regarding suppliers of the candidate items; a delivery route candidate generation unit that generates delivery route candidates when the item is delivered by the mobile body and the candidate items are loaded onto the mobile body based on the delivery destination information and the supplier information; an evaluation unit that evaluates the delivery route candidates based on at least one of the travel time, travel distance, and purchase probability information when the mobile object travels along the delivery route candidates, and determines the delivery route candidate whose evaluation result satisfies a predetermined condition as the delivery route along which the mobile object will travel; an output unit that outputs the delivery route; Information output device.

17. A program for causing a computer to execute the information output method according to any one of claims 1 to 3.

Images