Area determination method and information processing device

Abstract

To optimize assignment of delivery areas.SOLUTION: An area determination method for determining areas for delivery for each of multiple delivery bodies includes the following processes to be executed by a computer: an area candidate determination process to determine a plurality of area candidates; an extraction process to randomly extract a plurality of combinations of two demand points, based on the area candidates; a probability derivation process to derive probabilities that the delivery bodies conducted delivery, for each of the combinations, on the basis of past delivery data regarding deliveries to demand points by the delivery bodies; and an area determination process to determine areas on the basis of the probabilities.SELECTED DRAWING: Figure 5

Description

【Technical Field】 【0001】 The present invention relates to an area determination method and an information processing apparatus. 【Background Art】 【0002】 Conventionally, a LPG (Liquefied Petroleum Gas) container is provided to a consumer, and the replacement work of the LPG container is performed before the gas remaining amount of the LPG container becomes zero. For example, Patent Document 1 discloses detecting the gas usage amount or the gas remaining amount of the LPG container provided to the consumer, and calculating the replacement date of the LPG container based on the gas usage amount or the gas remaining amount. Further, Patent Document 1 discloses determining the number of delivery workers in the delivery area of the delivery vehicle that delivers the LPG container on the replacement date. 【Prior Art Documents】 【Patent Documents】 【0003】 【Patent Document 1】 Japanese Patent Application Laid-Open No. 2018-190253 【Summary of the Invention】 【Problems to be Solved by the Invention】 【0004】 By the way, a delivery area to be delivered is assigned to the delivery vehicle in advance. The distribution of consumers in the delivery area changes periodically. For example, an increase in consumers due to new construction or the like, or a decrease in consumers due to a move or the like can be considered. In addition, an increase or decrease in the usage amount of consumers due to a change in family composition or home time may affect the distribution of consumers who need to be delivered by the delivery vehicle during a certain period. Thus, as the distribution of consumers changes, the predetermined delivery area of the delivery vehicle may become inefficient. Therefore, when the distribution of consumers changes, it is necessary to review the assignment of the delivery area. Since an enormous amount of time is required for the work of reviewing the assignment of the delivery area, it has been difficult to optimize the assignment of the delivery area conventionally. 【0005】 The present invention aims to provide an area determination method and information processing device that can optimize the allocation of delivery areas. [Means for solving the problem] 【0007】 To solve the above problems, the present invention provides an area determination method for determining the delivery area for each of a plurality of delivery bodies, wherein a computer performs an area candidate determination process to determine a plurality of area candidates, performs an extraction process to randomly select a plurality of combinations of two demand points based on the area candidates, performs a probability derivation process to derive the probability that a delivery body delivered to each combination based on past delivery data of delivery bodies that delivered to demand points, and performs an area determination process to determine the area based on the probability. The extraction process randomly selects multiple combinations of two demand points within the same candidate area. The probability derivation process derives a first individual probability for each combination, which is the probability that the same delivery entity delivered the items. The first individual probabilities are then combined with multiple combinations to derive a first combined probability. The area determination process repeats the area candidate determination process, extraction process, and probability derivation process until the area candidate with the highest first combined probability is selected as the area. do . 【0008】 To solve the above problems, the present invention provides an area determination method for determining the delivery area for each of a plurality of delivery bodies, wherein a computer performs an area candidate determination process to determine a plurality of area candidates, performs an extraction process to randomly select a plurality of combinations of two demand points based on the area candidates, performs a probability derivation process to derive the probability that a delivery body delivered to each combination based on past delivery data of delivery bodies that delivered to demand points, and performs an area determination process to determine the area based on the probability. The extraction process randomly selects multiple combinations of two demand points from different area candidates. The probability derivation process derives a second individual probability for each combination, which is the probability that the same delivery entity delivered the items. These second individual probabilities are then combined to derive a second combined probability. The area determination process repeats the area candidate determination process, extraction process, and probability derivation process until the second combined probability is minimized. ru e Rear candidates by area do . 【0009】 To solve the above problems, the present invention provides an area determination method for determining the delivery area for each of a plurality of delivery bodies, wherein a computer performs an area candidate determination process to determine a plurality of area candidates, performs an extraction process to randomly select a plurality of combinations of two demand points based on the area candidates, performs a probability derivation process to derive the probability that a delivery body delivered to each combination based on past delivery data of delivery bodies that delivered to demand points, and performs an area determination process to determine the area based on the probability. The sampling process randomly selects multiple combinations of two demand points within the same area candidate. The probability derivation process derives a first individual probability for each combination selected from the same area candidate, which is the probability that the same delivery person delivered the items. The first individual probabilities are then combined with multiple combinations to derive a first combined probability. The sampling process randomly selects multiple combinations of two demand points within different area candidates. The probability derivation process derives a second individual probability for each combination selected from different area candidates, which is the probability that the same delivery person delivered the items. The second individual probabilities are then combined with multiple combinations to derive a second combined probability. The area determination process repeats the area candidate determination process, sampling process, and probability derivation process until the result obtained by subtracting the second combined probability from the first combined probability is maximized. ru e Rear candidate to E rear and do . 【0010】 To solve the above problems, the present invention provides an area determination method for determining the delivery area for each of a plurality of delivery bodies, wherein a computer performs an area candidate determination process to determine a plurality of area candidates, performs an extraction process to randomly select a plurality of combinations of two demand points based on the area candidates, performs a probability derivation process to derive the probability that a delivery body delivered to each combination based on past delivery data of delivery bodies that delivered to demand points, and performs an area determination process to determine the area based on the probability. The extraction process randomly selects multiple combinations of two demand points from different area candidates. The probability derivation process derives a third individual probability for each combination, which is the probability that a different delivery entity delivered the item. These third individual probabilities are then combined to derive a third combined probability. The area determination process repeats the area candidate determination process, extraction process, and probability derivation process until the area candidate with the highest third combined probability is selected as the area. do . 【0011】 To solve the above problems, the present invention provides an area determination method for determining the delivery area for each of a plurality of delivery bodies, wherein a computer performs an area candidate determination process to determine a plurality of area candidates, performs an extraction process to randomly select a plurality of combinations of two demand points based on the area candidates, performs a probability derivation process to derive the probability that a delivery body delivered to each combination based on past delivery data of delivery bodies that delivered to demand points, and performs an area determination process to determine the area based on the probability. The extraction process randomly selects multiple combinations of two demand points within the same area candidate. The probability derivation process derives a fourth individual probability for each combination, which is the probability that a different delivery body delivered the item. These fourth individual probabilities are then combined to derive a fourth combined probability. The area determination process repeats the area candidate determination process, extraction process, and probability derivation process until the area candidate with the smallest fourth combined probability is selected as the area. do . 【0012】 To solve the above problems, the present invention provides an area determination method for determining the delivery area for each of a plurality of delivery bodies, wherein a computer performs an area candidate determination process to determine a plurality of area candidates, performs an extraction process to randomly select a plurality of combinations of two demand points based on the area candidates, performs a probability derivation process to derive the probability that a delivery body delivered to each combination based on past delivery data of delivery bodies that delivered to demand points, and performs an area determination process to determine the area based on the probability. The sampling process randomly selects multiple combinations of two demand points from different area candidates. The probability derivation process derives a third individual probability for each combination selected from different area candidates, which is the probability that a different delivery body delivered the item. These third individual probabilities are then combined to derive a third combined probability. The sampling process randomly selects multiple combinations of two demand points from the same area candidate. The probability derivation process derives a fourth individual probability for each combination selected from the same area candidate, which is the probability that a different delivery body delivered the item. These fourth individual probabilities are then combined to derive a fourth combined probability. The area determination process repeats the area candidate determination process, sampling process, and probability derivation process until the area candidate that yields the maximum result of subtracting the fourth combined probability from the third combined probability is selected as the area. do . 【0013】 A set of demand points located within a predetermined range may be used instead of individual demand points. 【0014】 As the probability that the delivery entity has made a delivery, past performance values may be used. 【0015】 As the probability that the delivery entity has made a delivery, future predicted values may be used. 【0016】 In order to solve the above problems, an information processing apparatus of the present invention is an information processing apparatus including a computer that determines an area to be a delivery target for each of a plurality of delivery entities, and the computer includes: an area candidate determination process for determining a plurality of area candidates; an extraction process for randomly extracting a plurality of combinations of two demand points based on the area candidates; a probability derivation process for deriving the probability that the delivery entity has made a delivery for each combination based on past delivery data in which the delivery entity has made a delivery to the demand point; and an area determination process for determining an area based on the probability, and executes The extraction process randomly selects multiple combinations of two demand points within the same area candidate. The probability derivation process derives a first individual probability for each combination, which is the probability that the same delivery entity delivered the item. The first individual probabilities are then combined with multiple combinations to derive a first combined probability. The area determination process repeats the area candidate determination process, extraction process, and probability derivation process until the area candidate with the highest first combined probability is selected as the area. this. To solve the above problems, the present invention provides an information processing device equipped with a computer that determines the delivery area for each of a plurality of delivery bodies, wherein the computer performs an area candidate determination process to determine a plurality of area candidates, an extraction process to randomly extract a plurality of combinations of two demand points based on the area candidates, a probability derivation process to derive the probability that a delivery body delivered to each combination based on past delivery data of delivery bodies that have delivered to demand points, and an area determination process to determine an area based on the probability, wherein the extraction process randomly extracts a plurality of combinations of two demand points in different area candidates, the probability derivation process derives a second individual probability for each combination which is the probability that the same delivery body delivered, and the second individual probabilities are combined with a plurality of combinations to derive a second combined probability, and the area determination process repeats the area candidate determination process, the extraction process and the probability derivation process, and selects the area candidate that minimizes the second combined probability as the area. To solve the above problems, the present invention provides an information processing device equipped with a computer that determines the delivery area for each of a plurality of delivery bodies, wherein the computer performs an area candidate determination process that determines a plurality of area candidates, an extraction process that randomly extracts a plurality of combinations of two demand points based on the area candidates, a probability derivation process that derives the probability that a delivery body delivered to each combination based on past delivery data of delivery bodies that delivered to demand points, and an area determination process that determines an area based on the probability, wherein the extraction process randomly extracts a plurality of combinations of two demand points within the same area candidate, and the probability derivation process performs the same For each combination extracted from the area candidates, a first individual probability is derived, which is the probability that the same delivery body delivered the items. The first individual probability is then combined with multiple combinations to derive a first combined probability. The extraction process randomly selects multiple combinations of two demand points from different area candidates. The probability derivation process derives a second individual probability for each combination extracted from different area candidates, which is the probability that the same delivery body delivered the items. The second individual probability is then combined with multiple combinations to derive a second combined probability. The area determination process repeats the area candidate determination process, extraction process, and probability derivation process until the area candidate that yields the maximum result of subtracting the second combined probability from the first combined probability is selected as the area. To solve the above problems, the present invention provides an information processing device equipped with a computer that determines the delivery area for each of a plurality of delivery bodies, wherein the computer performs an area candidate determination process to determine a plurality of area candidates, an extraction process to randomly extract a plurality of combinations of two demand points based on the area candidates, a probability derivation process to derive the probability that a delivery body delivered to each combination based on past delivery data of delivery bodies that delivered to demand points, and an area determination process to determine an area based on the probability, wherein the extraction process randomly extracts a plurality of combinations of two demand points in different area candidates, the probability derivation process derives a third individual probability for each combination which is the probability that a different delivery body delivered, and the third individual probabilities are combined in a plurality of combinations to derive a third combined probability, and the area determination process repeats the area candidate determination process, the extraction process and the probability derivation process, and selects the area candidate that maximizes the third combined probability as the area. To solve the above problems, the present invention provides an information processing device equipped with a computer that determines the delivery area for each of a plurality of delivery bodies, wherein the computer performs an area candidate determination process to determine a plurality of area candidates, an extraction process to randomly extract a plurality of combinations of two demand points based on the area candidates, a probability derivation process to derive the probability that a delivery body delivered to each combination based on past delivery data of delivery bodies that delivered to demand points, and an area determination process to determine an area based on the probability, wherein the extraction process randomly extracts a plurality of combinations of two demand points in the same area candidate, the probability derivation process derives a fourth individual probability for each combination which is the probability that a different delivery body delivered, and the fourth individual probabilities are combined with a plurality of combinations to derive a fourth combined probability, and the area determination process repeats the area candidate determination process, the extraction process and the probability derivation process, and selects the area candidate that minimizes the fourth combined probability as the area. To solve the above problems, the present invention provides an information processing device equipped with a computer that determines the delivery area for each of a plurality of delivery bodies, wherein the computer performs an area candidate determination process that determines a plurality of area candidates, an extraction process that randomly extracts a plurality of combinations of two demand points based on the area candidates, a probability derivation process that derives the probability that a delivery body delivered to each combination based on past delivery data of delivery bodies that delivered to demand points, and an area determination process that determines an area based on the probability, wherein the extraction process randomly extracts a plurality of combinations of two demand points in different area candidates, and the probability derivation process randomly extracts a plurality of combinations of two demand points in different areas For each combination extracted from the area candidates, a third individual probability is derived, which is the probability that a different delivery body delivered the item. The third individual probability is then combined with multiple combinations of these third individual probabilities to derive a third combined probability. The extraction process randomly extracts multiple combinations of two demand points within the same area candidate. The probability derivation process derives a fourth individual probability for each combination extracted from the same area candidate, which is the probability that a different delivery body delivered the item. The fourth individual probability is then combined with multiple combinations of these fourth individual probabilities to derive a fourth combined probability. The area determination process repeats the area candidate determination process, extraction process, and probability derivation process until the area candidate that yields the maximum result of subtracting the fourth combined probability from the third combined probability is selected as the area. 【Effect of the Invention】 【0017】 According to the present invention, it is possible to optimize the allocation of the delivery area. 【Brief Description of the Drawings】 【0018】 [Figure 1] FIG. 1 is a schematic diagram showing a delivery system according to the present embodiment. [Figure 2] FIG. 2 is a functional block diagram for explaining the functions of the server according to the present embodiment. [Figure 3] FIG. 3 is a functional block diagram for explaining the functions of the delivery vehicle according to the present embodiment. [Figure 4] FIG. 4 is a diagram for explaining an example of an optimal delivery route that the delivery vehicle has delivered in the past. [Figure 5]Figure 5 is a flowchart illustrating the area determination method. [Figure 6] Figure 6 is a diagram illustrating an example of a candidate area. [Figure 7] Figure 7 is a diagram illustrating the multiple sub-regions included in each candidate area. [Modes for carrying out the invention] 【0019】 Preferred embodiments of the present invention will be described in detail below with reference to the attached drawings. The dimensions, materials, and other specific numerical values ​​shown in these embodiments are merely examples to facilitate understanding of the invention and do not limit the present invention unless otherwise specified. In this specification and drawings, elements having substantially the same function and configuration are denoted by the same reference numerals to avoid redundant explanations, and elements not directly related to the present invention are omitted from the illustrations. 【0020】 [1. Overall structure of the delivery system] First, the overall configuration of the delivery system 1000 according to this embodiment will be described with reference to Figure 1. Figure 1 is a schematic diagram showing the delivery system 1000 according to this embodiment. 【0021】 As shown in Figure 1, the delivery system 1000 according to this embodiment includes a server (information processing device) 100 and a plurality of delivery vehicles (delivery units) 200. The server 100 and the plurality of delivery vehicles 200 are connected to each other via a network NW so that they can communicate with one another. 【0022】 Server 100 is composed of various computer devices, such as server computers, personal computers, workstations, mainframe computers, and microcomputers. Server 100 collects, stores, and processes various information transmitted from multiple delivery vehicles 200. 【0023】 The delivery vehicle 200 transports fuel from the business site to the customer. Specifically, the delivery vehicle 200 transports LPG as fuel from the gas business site to the customer. The delivery vehicle 200 may, for example, supply LPG to LPG containers installed at the customer's site, or it may carry LPG containers filled with LPG and exchange the loaded LPG containers for LPG containers installed at the customer's site. As will be described in more detail later, the delivery vehicle 200 transmits various information such as its own location information, vehicle identification information, fuel supply information, and fuel level information to the server 100. 【0024】 The network NW is a wireless or wired communication network for communicating between the server 100 and multiple delivery vehicles 200. The network NW consists of various networks, such as satellite communication networks, mobile phone networks, the Internet, LANs (Local Area Networks), WANs (Wide Area Networks), and other dedicated network lines. At least a portion of the network NW includes a wireless network. However, the network NW may also include a wired network as a part. 【0025】 Figure 2 is a functional block diagram illustrating the functions of the server 100 according to this embodiment. As shown in Figure 2, the server 100 according to this embodiment includes a server communication unit 110, a server storage unit 130, and a server control unit 150. 【0026】 The server communication unit 110 establishes communication with delivery communication units 230 (see Figure 3) installed in multiple delivery vehicles 200 via the network NW. The server storage unit 130 consists of ROM, RAM, flash memory, HDD, etc., and stores programs and various data used by the server 100. 【0027】 The server control unit 150 is composed of a CPU (Central Processing Unit) and controls the entire server 100 using programs stored in the server memory unit 130. The server control unit 150 also functions as an information receiving unit 152, an area candidate determination processing unit 154, an extraction processing unit 156, a probability derivation unit 158, and an area determination processing unit 160. The probability derivation unit 158 ​​also functions as an individual probability derivation unit 158a and a combined probability derivation unit 158b. These functional units will be described in detail later. 【0028】 Figure 3 is a functional block diagram illustrating the functions of the delivery vehicle 200 according to this embodiment. As shown in Figure 3, the delivery vehicle 200 according to this embodiment has a delivery device 210. The delivery device 210 includes a delivery communication unit 230, a delivery storage unit 250, a GNSS (Global Navigation Satellite System) sensor 270, and a delivery control unit 290. 【0029】 The delivery communication unit 230 establishes communication with the server communication unit 110 (see Figure 2) installed on the server 100 via the network NW. The delivery storage unit 250 consists of ROM, RAM, flash memory, HDD, etc., and stores programs and various data used in the delivery device 210. The GNSS sensor 270 detects the position (latitude and longitude) of the delivery vehicle 200 and outputs a detection signal indicating the position of the delivery vehicle 200 to the delivery control unit 290. 【0030】 The delivery control unit 290 is composed of a CPU and controls the entire delivery device 210 using a program stored in the delivery memory unit 250. The delivery control unit 290 also functions as an information transmission unit 292 and a location information acquisition unit 294. These functional units will be described in detail later. 【0031】 Incidentally, each delivery vehicle (200) is assigned a predetermined delivery area. The distribution of customers within that delivery area changes periodically. For example, this could be due to an increase in customers due to new construction or a decrease due to relocation. Furthermore, changes in customer usage due to changes in family structure or time spent at home can affect the distribution of customers that delivery vehicle 200 needs to deliver to during a given period. As the distribution of customers changes in this way, the predetermined delivery area for delivery vehicle 200 may become inefficient. Therefore, if the distribution of customers changes, it is necessary to review the delivery area assignment. Reviewing the delivery area assignment requires a tremendous amount of time, making it difficult to optimize the delivery area assignment. 【0032】 Therefore, in this embodiment, the objective is to optimize the allocation of delivery areas based on past delivery data of deliveries made by the delivery vehicle 200 to customers. More specifically, in this embodiment, the objective is to optimize the allocation of delivery areas based on past delivery data of deliveries made by the delivery vehicle 200 to customers and derived data. Here, derived data is, for example, delivery data calculated by extracting the customer's available delivery dates from past delivery data, removing the concept of previously determined delivery areas, and ensuring that the delivery vehicle delivers to customers in a way that achieves a predetermined objective. The predetermined objective is, for example, to minimize the delivery distance of the delivery vehicle or to minimize the travel time of the delivery vehicle. Also, derived data is, for example, delivery data calculated by virtually creating the customer's available delivery dates from current or future delivery data, and ensuring that the delivery vehicle delivers to customers in a way that achieves a predetermined objective. However, when creating the virtual customer's available delivery dates, past delivery data may be used for estimation. In the following section, the optimal delivery area for delivery vehicle 200 is estimated (maximum likelihood estimation of the delivery area) based on delivery data including the optimal delivery route previously used by delivery vehicle 200, and the derived data described above. The area determination method for maximum likelihood estimation of the optimal delivery area for delivery vehicle 200 in this embodiment is described below. 【0033】 Figure 4 illustrates an example of an optimal delivery route previously used by delivery vehicle 200. In Figure 4, P1 is the location of the business establishment where delivery vehicle 200 begins its delivery (hereinafter referred to as the facility point), and P2 is the location of the customer (hereinafter referred to as the demand point). In Figure 4, all points other than the facility point P1 are demand points P2. 【0034】 As shown in Figure 4, delivery vehicle 200A, one of the multiple delivery vehicles 200, transports fuel to each demand point P2 along delivery route A, shown by the solid line. Delivery vehicle 200B, another of the multiple delivery vehicles 200, transports fuel to each demand point P2 along delivery route B, shown by the dashed line. Delivery vehicle 200C, another of the multiple delivery vehicles 200, transports fuel to each demand point P2 along delivery route C, shown by the dashed line. In Figure 4, the demand points P2 not located on delivery routes A, B, and C represent the locations of customers who do not yet require fuel delivery, or the locations of customers on days when fuel delivery is not possible. 【0035】 At this time, in order to construct a database to be referenced when making the maximum likelihood estimation of the delivery area, the location information acquisition unit 294 of the delivery device 210 mounted on each delivery vehicle 200 acquires location information from the GNSS sensor 270. In addition, the delivery storage unit 250 stores vehicle identification information. Vehicle identification information is provided for each delivery vehicle 200A, 200B, and 200C and is information for distinguishing between its own vehicle and other vehicles. The vehicle identification information may also include information regarding the maximum load capacity of LPG or LPG containers that the delivery vehicle 200 can carry. 【0036】 The information transmission unit 292 of the delivery device 210 transmits the location information of the delivery vehicle 200 acquired by the location information acquisition unit 294, the time information, and the vehicle identification information stored in the delivery storage unit 250 to the server communication unit 110 via the delivery communication unit 230. Here, the information transmission unit 292 transmits the location information, time information, and vehicle identification information of the delivery vehicle 200 to the server communication unit 110 when the delivery vehicle 200 arrives at the demand point P2 and supplies LPG or LPG containers to the demand point P2. 【0037】 In this case, the information transmission unit 292 may transmit fuel supply information to the server communication unit 110, along with location information, time information, and vehicle identification information, regarding the amount of LPG to be supplied to the LPG container at demand point P2 or the amount of LPG containers to be replaced (number of containers to be replaced). The information transmission unit 292 may also transmit fuel remaining amount information to the server communication unit 110, regarding the remaining amount of LPG loaded on the delivery vehicle 200 or the number of remaining LPG containers. 【0038】 The information receiving unit 152 of the server 100 acquires various information about the delivery vehicle 200 from the information transmitting unit 292 of the delivery device 210. Once the information receiving unit 152 acquires various information from the delivery vehicle 200, it stores the acquired information in the server storage unit 130. The server storage unit 130 stores various information about the delivery vehicle 200. 【0039】 The area determination processing unit 160 of the server 100 performs maximum likelihood estimation of the delivery area of ​​the delivery vehicle 200 based on various information of the delivery vehicle 200 stored in the server storage unit 130. The area determination method for maximum likelihood estimation of the delivery area will be described below. 【0040】 Figure 5 is a flowchart of the area determination method. Although Figure 5 shows the processing flow of the area determination method in the first embodiment, the basic processing flow is the same in the second and third embodiments, so the explanation of the content that overlaps with the first embodiment in the second and third embodiments will be omitted. As shown in Figure 5, in order to estimate the delivery area with maximum likelihood, first, the area candidate determination processing unit 154 of the server 100 performs an area candidate determination process to determine the area candidates for the delivery area to be delivered by each of the multiple delivery vehicles 200 (S100). 【0041】 Figure 6 illustrates an example of area candidates. In Figure 6, three area candidates Ra, Rb, and Rc are shown, demarcated by dashed lines. Area candidate Ra is the delivery area to be served by delivery vehicle 200A, area candidate Rb is the delivery area to be served by delivery vehicle 200B, and area candidate Rc is the delivery area to be served by delivery vehicle 200C. Each area candidate includes at least one demand point P2. 【0042】 The number of area candidates is determined by the number of delivery vehicles 200. For example, in Figure 6, there are three delivery vehicles 200A, 200B, and 200C, so there are three area candidates: area candidate Ra, area candidate Rb, and area candidate Rc. In other words, the number of area candidates is equal to the number of delivery vehicles 200. 【0043】 The area candidate determination processing unit 154 determines each area candidate such that area candidates Ra, Rb, and Rc do not overlap. Here, each demand point P2 is set to belong to at least one area candidate. Therefore, it is also possible for each demand point P2 to be set to belong to multiple area candidates. In that case, each area candidate may be set to overlap with one another. The areas of each area candidate may be equal or different. For example, the area of ​​each area candidate may differ depending on the distance from the facility point P1. Specifically, the area of ​​each area candidate may become smaller as the distance from the facility point P1 increases. 【0044】 Furthermore, each area candidate is set up in a way that balances the number of demand points P2 within each area candidate and the amount of demand within each area candidate to remain constant. For example, each area candidate is set up so that the difference in the number of demand points P2 within each area candidate is less than a threshold. Also, each area candidate is set up so that the difference in the amount of demand within each area candidate is less than a threshold. However, this is not limited to the above, and the number of demand points P2 and the amount of demand included in each area candidate may differ depending on, for example, the distance from facility point P1. Specifically, the number of demand points P2 and the amount of demand included in each area candidate may decrease as the distance from facility point P1 increases. 【0045】 Furthermore, there may be limitations on the amount of fuel that can be carried in the delivery vehicle 200. Also, there may be limitations on the number of demand points P2 that one delivery person can visit on the delivery route. Similarly, there may be limitations on the delivery time that one delivery person can spend on the delivery route. Thus, there may be limitations on at least one of the following: the amount of fuel that can be carried in the delivery vehicle 200, the number of demand points P2 on the delivery route, or the delivery time, such as only the fuel capacity, only the number of demand points P2, or only the delivery time. Therefore, the area candidate determination processing unit 154 determines the area candidates by taking these limitations (the amount of fuel that can be carried in the delivery vehicle 200, the number of demand points P2 on the delivery route, delivery time, etc.) into consideration. 【0046】 The area candidate determination processing unit 154 creates multiple patterns of area candidates Ra, Rb, and Rc. Each of the multiple patterns of area candidates Ra, Rb, and Rc has a different shape and size. The area candidate determination processing unit 154 selectively determines one area candidate Ra, Rb, or Rc from among the multiple patterns of area candidates Ra, Rb, and Rc that have been created. 【0047】 Next, the area candidate determination processing unit 154 performs a process to derive area candidates Ra, Rb, and Rc from among the multiple patterns of area candidates Ra, Rb, and Rc created, such that at least one of the following first and second points is achieved. 【0048】 The first point here is to "maximize the probability that two demand points P2 within the same candidate area are delivered by the same delivery vehicle 200, when they are randomly selected." The second point is to "minimize the probability that two demand points P2 within different candidate areas are delivered by the same delivery vehicle 200, when they are randomly selected." 【0049】 First, in order to satisfy the first point condition, the extraction processing unit 156 performs an extraction process to randomly select multiple combinations of two demand points P2 within the same candidate area (S200). 【0050】 Next, the individual probability derivation unit 158a performs an individual probability derivation process (S300) to derive an individual probability (first individual probability), which is the probability that the same delivery vehicle 200 delivered the goods for each combination of two demand points P2 extracted from the same area candidate. 【0051】 Furthermore, the combined probability derivation unit 158b performs a combined probability derivation process (S400) in which the individual probabilities (first individual probabilities) derived by the individual probability derivation unit 158a are combined in multiple combinations to derive a combined probability (first combined probability). 【0052】 The area determination processing unit 160 then repeatedly executes the area candidate determination process, extraction process, individual probability derivation process, and combined probability derivation process described above, and determines the area candidate with the highest first combined probability based on the following formula (1) as the delivery area to be delivered by each of the multiple delivery vehicles 200 (S500). In this embodiment, the delivery area is determined using the following formula (1), but the formula used to determine the delivery area is not limited to this. For example, the delivery area may be determined using an approximate formula of the following formula (1). 【number】 【0053】 Here, i,j (∈I) are indices of two demand points P2. a is an index of area candidates. k is an index of trials in which the area candidate determination process, extraction process, individual probability derivation process, and composite probability derivation process are repeated. I is the set of demand points P2. A is the set of area candidates. K is the set of trials. k is the set of demand points P2 visited by trial k. C(X) is the definition of a set→set function that generates a set of all possible combinations of selecting two demand points P2 from a given set of points X. TIFF0007873799000002.tif31170 is a parameter that is 1 when demand points i and j are delivered by the same delivery vehicle 200 in trial k, and 0 otherwise. ia This is a decision variable that is 1 when demand point i belongs to area a, and 0 otherwise. 【0054】 According to the delivery area determined by the area determination processing unit 160, based on past delivery data, two demand points P2 belonging to the same area will be delivered by the same delivery vehicle 200 as much as possible. By using the above formula (1) to determine the delivery area, the allocation of delivery areas can be easily optimized and its optimality can be improved. In addition, the automation of the allocation of delivery areas reduces the burden on workers. 【0055】 Furthermore, the delivery area may be determined not by the condition of the first point, but by satisfying the condition of the second point. In order to satisfy the condition of the second point, the extraction processing unit 156 performs an extraction process to randomly select multiple combinations of two demand points P2 in different area candidates (S200). 【0056】 Next, the individual probability derivation unit 158a performs an individual probability derivation process (S300) to derive an individual probability (second individual probability), which is the probability that the same delivery vehicle 200 delivered the items, for each combination of two demand points P2 extracted from different area candidates. 【0057】 Furthermore, the combined probability derivation unit 158b performs a combined probability derivation process (S400) in which it combines the individual probabilities (second individual probabilities) derived by the individual probability derivation unit 158a in multiple combinations to derive a combined probability (second combined probability). 【0058】 The area determination processing unit 160 then repeatedly executes the area candidate determination process, extraction process, individual probability derivation process, and combined probability derivation process described above, and determines the area candidate that minimizes the second combined probability based on the following formula (2) as the delivery area to be delivered by each of the multiple delivery vehicles 200 (S500). In this embodiment, the delivery area is determined using the following formula (2), but the formula used to determine the delivery area is not limited to this. For example, the delivery area may be determined using an approximate formula of the following formula (2). 【number】 【0059】 According to the delivery area determined by the area determination processing unit 160, based on past delivery data, two demand points P2 belonging to different areas will be delivered by different delivery vehicles 200 whenever possible. By using the above formula (2) to determine the delivery area, the allocation of delivery areas can be easily optimized and its optimality can be increased. Furthermore, automating the allocation of delivery areas reduces the burden on workers. 【0060】 Furthermore, the determination of the delivery area may be based on two perspectives other than the first and second points, or in addition, on both perspectives simultaneously. For example, the delivery area may be determined by satisfying both the conditions of the first point and the conditions of the second point at the same time. Specifically, the area determination processing unit 160 determines the area candidate that maximizes the result of subtracting the second combined probability from the first combined probability, based on the above formulas (1) and (2), as the delivery area to be served by each of the multiple delivery vehicles 200 (S500). 【0061】 According to the delivery area determined by the area determination processing unit 160, based on past delivery data, two demand points P2 belonging to the same area will be delivered by the same delivery vehicle 200 as much as possible, and two demand points P2 belonging to different areas will be delivered by different delivery vehicles 200 as much as possible. By using the above formulas (1) and (2) to determine the delivery area, the allocation of delivery areas can be easily optimized and its optimality can be increased. In addition, the workload on workers can be reduced by automating the allocation of delivery areas. 【0062】 (Second Embodiment) If the volume of past delivery data or the number of demand points P2 is large, the calculation process for maximum likelihood estimation in the first embodiment may take a long time. Also, in certain areas (e.g., XX block, YY address, etc.), it may be more efficient to deliver using the same delivery vehicle 200. Therefore, in the second embodiment, each area candidate Ra, Rb, and Rc is divided into multiple sub-regions SA. 【0063】 Figure 7 is a diagram illustrating the multiple sub-regions SA contained within each candidate area Ra, Rb, and Rc. As shown in Figure 7, each candidate area Ra, Rb, and Rc consists of multiple sub-regions SA. Each sub-region SA contains at least one demand point P2. In the second embodiment, each sub-region SA is a specific area, for example, a block or address. 【0064】 Each sub-region SA is set up in a balanced manner so that the number of demand points P2 within each sub-region SA and the amount of demand within each sub-region SA remain constant. Specifically, each sub-region SA is set up so that the difference in the number of demand points P2 within each sub-region SA and the amount of demand within each sub-region SA is below a threshold. 【0065】 In the first embodiment, two demand points P2 within a candidate area are randomly selected, while in the second embodiment, two sub-regions SA within the candidate area are randomly selected. In other words, in the second embodiment, multiple demand points P2 located within a predetermined range within the candidate area are treated as sub-regions (demand sets), and the maximum likelihood estimation calculation is performed using these sub-regions (demand sets) instead of the demand points P2. 【0066】 In the maximum likelihood estimation calculation process in the second embodiment, a process is performed to derive area candidate Ra, area candidate Rb, and area candidate Rc that satisfy at least one of the following third and fourth points. 【0067】 The third point here is to "maximize the probability that two sub-regions SA within the same candidate area are delivered by the same delivery vehicle 200 when randomly selected." The fourth point is to "minimize the probability that two sub-regions SA within different candidate areas are delivered by the same delivery vehicle 200 when randomly selected." 【0068】 First, in order to satisfy the third point condition, the extraction processing unit 156 performs an extraction process to randomly select multiple combinations of two sub-regions SA within the same candidate area (S200). 【0069】 Next, the individual probability derivation unit 158a performs an individual probability derivation process (S300) to derive an individual probability (first individual probability), which is the probability that the same delivery vehicle 200 delivered the goods for each combination of two sub-regions SA extracted from the same area candidate. 【0070】 Furthermore, the combined probability derivation unit 158b performs a combined probability derivation process (S400) in which the individual probabilities (first individual probabilities) derived by the individual probability derivation unit 158a are combined in multiple combinations to derive a combined probability (first combined probability). 【0071】 The area determination processing unit 160 then repeatedly executes area candidate determination processing, extraction processing, individual probability derivation processing, and composite probability derivation processing, and determines the area candidate with the highest first composite probability based on the following formula (3) as the delivery area to be delivered by each of the multiple delivery vehicles 200 (S500). In this embodiment, the delivery area is determined using the following formula (3), but the formula used to determine the delivery area is not limited to this. For example, the delivery area may be determined using an approximate formula of the following formula (3). 【number】 【0072】 Here, p and q are indices of two sub-regions. a is an index of a candidate area. k is an index of a trial in which the area candidate determination process, extraction process, individual probability derivation process, and composite probability derivation process are repeated. P is the set of sub-regions SA. A is the set of candidate areas. K is the set of trials. C(X) is the definition of a set→set function that generates a set of all combinations of selecting two sub-regions SA from a given point set X. TIFF0007873799000005.tif24170 represents demand point I visited in trial k. k All combinations C(I k This parameter indicates the number of combinations in which one demand point is located in sub-region p, the other in sub-region q, and both are delivered by the same delivery vehicle 200. TIFF0007873799000006.tif29170 represents demand point I visited in trial k. k All combinations C(I k This parameter indicates the number of combinations in which one demand point is in sub-region p and the other is in sub-region q. pa This is a decision variable that is 1 when the sub-region p belongs to area a, and 0 otherwise. 【0073】 According to the second embodiment, the same functions and effects as the first embodiment can be achieved. Furthermore, according to the second embodiment, the maximum likelihood estimation calculation can be performed in a shorter time compared to the first embodiment. 【0074】 Furthermore, the delivery area may be determined not by the condition of the third point, but by satisfying the condition of the fourth point. In order to satisfy the condition of the fourth point, the extraction processing unit 156 performs an extraction process to randomly select multiple combinations of two sub-regions SA in different area candidates (S200). 【0075】 Next, the individual probability derivation unit 158a performs an individual probability derivation process (S300) to derive an individual probability (second individual probability), which is the probability that the same delivery vehicle 200 delivered the goods, for each combination of two sub-regions SA extracted from different area candidates. 【0076】 Furthermore, the combined probability derivation unit 158b performs a combined probability derivation process (S400) in which the individual probabilities (second individual probabilities) derived by the individual probability derivation unit 158a are combined in multiple combinations to derive a combined probability (second combined probability). 【0077】 The area determination processing unit 160 then repeatedly executes area candidate determination processing, extraction processing, individual probability derivation processing, and composite probability derivation processing, and determines the area candidate that minimizes the second composite probability based on the following formula (4) as the delivery area to be delivered by each of the multiple delivery vehicles 200 (S500). In this embodiment, the delivery area is determined using the following formula (4), but the formula used to determine the delivery area is not limited to this. For example, the delivery area may be determined using an approximate formula of the following formula (4). 【number】 【0078】 According to the second embodiment, the same functions and effects as the first embodiment can be achieved. Furthermore, according to the second embodiment, the maximum likelihood estimation calculation can be performed in a shorter time compared to the first embodiment. 【0079】 Furthermore, the determination of the delivery area may be based on, or in addition to, the two perspectives of the third and fourth points, or by considering both perspectives simultaneously. For example, the delivery area may be determined by satisfying both the conditions of the third point and the conditions of the fourth point at the same time. Specifically, the area determination processing unit 160 determines the area candidate that maximizes the result of subtracting the second combined probability from the first combined probability, based on the above formulas (3) and (4), as the area to be delivered to each of the multiple delivery vehicles 200. 【0080】 According to the second embodiment, the same functions and effects as the first embodiment can be achieved. Furthermore, according to the second embodiment, the maximum likelihood estimation calculation can be performed in a shorter time compared to the first embodiment. 【0081】 (Third embodiment) In the first embodiment, a process was performed to derive candidate areas Ra, Rb, and Rc that satisfy at least one of the first and second points described above. 【0082】 Here, the first point of the first embodiment is equivalent to finding a solution that maximizes the probability that two demand points P2 within different candidate areas are delivered by different delivery vehicles 200 (the fifth point). 【0083】 Furthermore, the second point in the first embodiment is equivalent to finding a solution that minimizes the probability that two demand points P2 within the same candidate area are delivered by different delivery vehicles 200 (the sixth point). 【0084】 Therefore, in the third embodiment, the area candidate determination processing unit 154 performs a process to derive area candidates Ra, Rb, and Rc from among a plurality of area candidate patterns Ra, Rb, and Rc that satisfy at least one of the fifth point and the sixth point. 【0085】 First, in order to satisfy the fifth point condition, the extraction processing unit 156 performs an extraction process to randomly select multiple combinations of two demand points P2 in different area candidates (S200). 【0086】 Next, the individual probability derivation unit 158a performs an individual probability derivation process (S300) to derive an individual probability (third individual probability), which is the probability that a different delivery vehicle 200 delivered the goods, for each combination of two demand points P2 extracted from different area candidates. 【0087】 Furthermore, the combined probability derivation unit 158b performs a combined probability derivation process (S400) in which the individual probabilities (third individual probabilities) derived by the individual probability derivation unit 158a are combined in multiple combinations to derive a combined probability (third combined probability). 【0088】 The area determination processing unit 160 then repeatedly executes the area candidate determination process, extraction process, individual probability derivation process, and composite probability derivation process, and based on the above formula (1), determines the area candidate that maximizes the third composite probability as the delivery area to be delivered by each of the multiple delivery vehicles 200 (S500). 【0089】 According to the third embodiment, the same actions and effects as the first embodiment can be achieved. 【0090】 Furthermore, the delivery area may be determined not by the condition of the fifth point, but by satisfying the condition of the sixth point. In order to satisfy the condition of the sixth point, the extraction processing unit 156 performs an extraction process to randomly select multiple combinations of two demand points P2 within the same candidate area (S200). 【0091】 Next, the individual probability derivation unit 158a performs an individual probability derivation process (S300) to derive an individual probability (fourth individual probability), which is the probability that a different delivery vehicle 200 delivered the goods, for each combination extracted from the same area candidate. 【0092】 Furthermore, the combined probability derivation unit 158b performs a combined probability derivation process (S400) in which the individual probabilities (fourth individual probabilities) derived by the individual probability derivation unit 158a are combined in multiple combinations to derive a combined probability (fourth combined probability). 【0093】 The area determination processing unit 160 then repeatedly executes area candidate determination processing, extraction processing, individual probability derivation processing, and composite probability derivation processing, and based on the above formula (2), derives the determination of the area candidate that minimizes the fourth composite probability as the delivery area to be delivered by each of the multiple delivery vehicles 200 (S500). 【0094】 According to the third embodiment, the same actions and effects as the first embodiment can be achieved. 【0095】 Furthermore, the determination of the delivery area may be based on, or in addition to, the two perspectives of the fifth and sixth points, or by considering both perspectives simultaneously. For example, the delivery area may be determined by simultaneously satisfying both the conditions of the fifth point and the conditions of the sixth point. Specifically, the area determination processing unit 160 determines the area candidate that maximizes the result of subtracting the fourth combined probability from the third combined probability, based on the above formulas (1) and (2), as the area to be delivered to each of the multiple delivery vehicles 200. 【0096】 According to the third embodiment, the same functions and effects as the first embodiment can be achieved. While this description has shown an example where the third embodiment is applied to the first embodiment, it is not limited to this example, and the third embodiment may also be applied to the second embodiment. 【0097】 As described above, according to the first to third embodiments, the optimal delivery area for the delivery vehicle 200 is estimated with maximum likelihood based on various information about the delivery vehicle 200 stored in the server storage unit 130. Specifically, the extraction processing unit 156 performs an extraction process to randomly select multiple combinations of two demand points P2 (or two sub-regions SA) based on the area candidate Ra, Rb, and Rc determined by the area candidate determination processing unit 154. The probability derivation unit 158 ​​performs a probability derivation process to derive the probability that the delivery vehicle 200 delivered to each combination of two demand points P2 (or two sub-regions SA) extracted by the extraction processing unit 156, based on past delivery data of the delivery vehicle 200 having delivered to demand points P2. Then, the area determination processing unit 160 performs an area determination process to determine the areas to be delivered to by each of the multiple delivery vehicles 200 based on the probabilities derived by the probability derivation unit 158. This makes it easy to optimize the allocation of delivery areas. 【0098】 Preferred embodiments of the present invention have been described above with reference to the attached drawings, but it goes without saying that the present invention is not limited to these embodiments. It is clear to those skilled in the art that various modifications or alterations can be conceived within the scope of the claims, and these will naturally also fall within the technical scope of the present invention. 【0099】 In the first to third embodiments described above, the probability that the delivery vehicle 200 made a delivery was derived using past delivery data. However, the invention is not limited to this, and the probability that the delivery vehicle 200 made a delivery may also be derived using future predicted values ​​derived based on past delivery data. 【0100】 In the first to third embodiments described above, an example was given in which the delivery vehicle 200 delivers fuel (LPG). However, the delivery vehicle 200 can be any vehicle that travels between a business establishment and a customer, and is not limited to one that delivers fuel. For example, it could be a vehicle that delivers goods or a vehicle that visits customers. [Explanation of symbols] 【0101】 P1 Facility Point P2 Each demand point Ra Area Candidates Rb Area Candidates Rc Area Candidates SA small area 100 servers 110 Server Communication Unit 130 Server Storage Unit 150 Server Control Unit 152 Information Receiving Unit 154 Area Candidate Determination Processing Unit 156 Extraction Processing Unit 158 Probability Derivation Section 158a Individual Probability Derivation Unit 158b Composite Probability Derivation Unit 160 Area Determination Processing Unit 200 delivery vehicle 210 Delivery device 230 Delivery and Communications Department 250 Delivery storage section 270 sensors 290 Delivery Control Unit 292 Information Transmission Section 294 Location information acquisition unit 1000 Delivery Systems

Claims

[Claim 1] An area determination method for determining the area to which each of a plurality of delivery bodies will be delivered, Computers The area candidate determination process is executed to determine multiple area candidates. Based on the aforementioned area candidates, an extraction process is performed to randomly select multiple combinations of two demand points. Based on past delivery data of the delivery entity that has delivered to the demand point, a probability derivation process is performed to derive the probability that the delivery entity delivered for each combination. Based on the aforementioned probability, an area determination process is performed to determine the area. The extraction process described above is: Within the same candidate area, multiple combinations of two demand points are randomly selected. The aforementioned probability derivation process is, For each of the above combinations, a first individual probability is derived, which is the probability that the same delivery body delivered the item. The first individual probabilities are combined with a plurality of the above combinations to derive the first combined probability. The area determination process described above is: The area candidate determination process, the extraction process, and the probability derivation process are repeated. The area candidate that maximizes the first combination probability is defined as the area. How to determine the area. [Claim 2] A method for determining the area to which each of a plurality of delivery bodies will be delivered, Computers The area candidate determination process is executed to determine multiple area candidates. Based on the aforementioned area candidates, an extraction process is performed to randomly select multiple combinations of two demand points. Based on past delivery data of the delivery entity that has delivered to the demand point, a probability derivation process is performed to derive the probability that the delivery entity delivered for each combination. Based on the aforementioned probability, an area determination process is performed to determine the area. The extraction process described above is: In the different candidate areas, multiple combinations of two demand points are randomly selected. The aforementioned probability derivation process is, For each of the above combinations, a second individual probability is derived, which is the probability that the same delivery body delivered the item. The second individual probabilities are combined with multiple combinations of the above to derive a second combined probability. The area determination process described above is: The area candidate determination process, the extraction process, and the probability derivation process are repeated. The area candidate that minimizes the second combination probability is defined as the area. How to determine the area. [Claim 3] An area determination method for determining the area to which each of a plurality of delivery bodies will be delivered, Computers The area candidate determination process is executed to determine multiple area candidates. Based on the aforementioned area candidates, an extraction process is performed to randomly select multiple combinations of two demand points. Based on past delivery data of the delivery entity that has delivered to the demand point, a probability derivation process is performed to derive the probability that the delivery entity delivered for each combination. Based on the aforementioned probability, an area determination process is performed to determine the area. The extraction process described above is: Within the same candidate area, multiple combinations of two demand points are randomly selected. The aforementioned probability derivation process is, For each combination extracted from the same area candidate, a first individual probability is derived, which is the probability that the same delivery entity delivered the items. The first individual probabilities are combined with a plurality of the above combinations to derive the first combined probability. The extraction process described above is: In the different candidate areas, multiple combinations of two demand points are randomly selected. The aforementioned probability derivation process is, For each combination extracted from the different area candidates, a second individual probability is derived, which is the probability that the same delivery entity delivered the items. The second individual probabilities are combined with multiple combinations of the above to derive a second combined probability. The area determination process described above is: The area candidate determination process, the extraction process, and the probability derivation process are repeated. The area candidate that yields the maximum result when the second combination probability is subtracted from the first combination probability is defined as the area. How to determine the area. [Claim 4] An area determination method for determining the area to which each of a plurality of delivery bodies will be delivered, Computers The area candidate determination process is executed to determine multiple area candidates. Based on the aforementioned area candidates, an extraction process is performed to randomly select multiple combinations of two demand points. Based on past delivery data of the delivery entity that has delivered to the demand point, a probability derivation process is performed to derive the probability that the delivery entity delivered for each combination. Based on the aforementioned probability, an area determination process is performed to determine the area. The extraction process described above is: In the different candidate areas, multiple combinations of two demand points are randomly selected. The aforementioned probability derivation process is, For each of the above combinations, a third individual probability is derived, which is the probability that a different delivery body was delivered. The third individual probabilities are combined with multiple combinations of the above to derive a third combined probability. The area determination process described above is: The area candidate determination process, the extraction process, and the probability derivation process are repeated. The area candidate that maximizes the third combination probability is defined as the area. How to determine the area. [Claim 5] An area determination method for determining the area to which each of a plurality of delivery bodies will be delivered, Computers The area candidate determination process is executed to determine multiple area candidates. Based on the aforementioned area candidates, an extraction process is performed to randomly select multiple combinations of two demand points. Based on past delivery data of the delivery entity that has delivered to the demand point, a probability derivation process is performed to derive the probability that the delivery entity delivered for each combination. Based on the aforementioned probability, an area determination process is performed to determine the area. The extraction process described above is: Within the same candidate area, multiple combinations of two demand points are randomly selected. The aforementioned probability derivation process is, For each of the aforementioned combinations, a fourth individual probability is derived, which is the probability that a different delivery body was delivered. The fourth individual probability is combined with a plurality of the above combinations to derive the fourth combined probability. The area determination process described above is: The area candidate determination process, the extraction process, and the probability derivation process are repeated. The area candidate that minimizes the fourth combination probability is defined as the area. How to determine the area. [Claim 6] An area determination method for determining the area to which each of a plurality of delivery bodies will be delivered, Computers The area candidate determination process is executed to determine multiple area candidates. Based on the aforementioned area candidates, an extraction process is performed to randomly select multiple combinations of two demand points. Based on past delivery data of the delivery entity that has delivered to the demand point, a probability derivation process is performed to derive the probability that the delivery entity delivered for each combination. Based on the aforementioned probability, an area determination process is performed to determine the area. The extraction process described above is: In the different candidate areas, multiple combinations of two demand points are randomly selected. The aforementioned probability derivation process is, For each combination extracted from the different area candidates, a third individual probability is derived, which is the probability that a different delivery entity delivered the package. The third individual probabilities are combined with multiple combinations of the above to derive a third combined probability. The extraction process described above is: Within the same candidate area, multiple combinations of two demand points are randomly selected. The aforementioned probability derivation process is, For each combination extracted from the same area candidate, a fourth individual probability is derived, which is the probability that a different delivery entity delivered the package. The fourth individual probability is combined with a plurality of the above combinations to derive the fourth combined probability. The area determination process described above is: The area candidate determination process, the extraction process, and the probability derivation process are repeated. The area candidate that yields the maximum result when the third combination probability is subtracted from the fourth combination probability is defined as the area. How to determine the area. [Claim 7] Multiple demand points located within a predetermined range are defined as a demand set. The area determination method according to any one of claims 1 to 6, which uses a set of demand points instead of the aforementioned demand points. [Claim 8] The area determination method according to any one of claims 1 to 6, wherein past performance values ​​are used as the probability that the delivery body made the delivery. [Claim 9] The area determination method according to any one of claims 1 to 6, wherein a future predicted value is used as the probability that the delivery body has made a delivery. [Claim 10] An information processing device equipped with a computer that determines the delivery area for each of multiple delivery units, The aforementioned computer, Area candidate determination process that determines multiple area candidates, Based on the aforementioned area candidates, an extraction process is performed to randomly select multiple combinations of two demand points, A probability derivation process that derives the probability that the delivery body delivered to the demand point for each combination, based on past delivery data of the delivery body that delivered to the demand point. An area determination process that determines the area based on the aforementioned probability, Execute, The extraction process described above is: Within the same candidate area, multiple combinations of two demand points are randomly selected. The aforementioned probability derivation process is, For each of the above combinations, a first individual probability is derived, which is the probability that the same delivery body delivered the item. The first individual probabilities are combined with a plurality of the above combinations to derive the first combined probability. The area determination process described above is: The area candidate determination process, the extraction process, and the probability derivation process are repeated. The area candidate that maximizes the first combination probability is defined as the area. Information processing device. [Claim 11] An information processing device comprising a computer that determines the delivery area for each of a plurality of delivery bodies, The aforementioned computer, Area candidate determination process that determines multiple area candidates, Based on the aforementioned area candidates, an extraction process is performed to randomly select multiple combinations of two demand points, A probability derivation process that derives the probability that the delivery body delivered to the demand point for each combination, based on past delivery data of the delivery body that delivered to the demand point. An area determination process that determines the area based on the aforementioned probability, Execute, The extraction process described above is: In the different candidate areas, multiple combinations of two demand points are randomly selected. The aforementioned probability derivation process is, For each of the above combinations, a second individual probability is derived, which is the probability that the same delivery body delivered the item. The second individual probabilities are combined with multiple combinations of the above to derive a second combined probability. The area determination process described above is: The area candidate determination process, the extraction process, and the probability derivation process are repeated. The area candidate that minimizes the second combination probability is defined as the area. Information processing device. [Claim 12] An information processing device comprising a computer that determines the delivery area for each of a plurality of delivery bodies, The aforementioned computer, Area candidate determination process that determines multiple area candidates, Based on the aforementioned area candidates, an extraction process is performed to randomly select multiple combinations of two demand points, A probability derivation process that derives the probability that the delivery body delivered to the demand point for each combination, based on past delivery data of the delivery body that delivered to the demand point. An area determination process that determines the area based on the aforementioned probability, Execute, The extraction process described above is: Within the same candidate area, multiple combinations of two demand points are randomly selected. The aforementioned probability derivation process is, For each combination extracted from the same area candidate, a first individual probability is derived, which is the probability that the same delivery entity delivered the items. The first individual probabilities are combined with a plurality of the above combinations to derive the first combined probability. The extraction process described above is: In the different candidate areas, multiple combinations of two demand points are randomly selected. The aforementioned probability derivation process is, For each combination extracted from the different area candidates, a second individual probability is derived, which is the probability that the same delivery entity delivered the items. The second individual probabilities are combined with multiple combinations of the above to derive a second combined probability. The area determination process described above is: The area candidate determination process, the extraction process, and the probability derivation process are repeated. The area candidate that yields the maximum result when the second combination probability is subtracted from the first combination probability is defined as the area. Information processing device. [Claim 13] An information processing device comprising a computer that determines the delivery area for each of a plurality of delivery bodies, The aforementioned computer, Area candidate determination process that determines multiple area candidates, Based on the aforementioned area candidates, an extraction process is performed to randomly select multiple combinations of two demand points, A probability derivation process that derives the probability that the delivery body delivered to the demand point for each combination, based on past delivery data of the delivery body that delivered to the demand point. An area determination process that determines the area based on the aforementioned probability, Execute, The extraction process described above is: In the different candidate areas, multiple combinations of two demand points are randomly selected. The aforementioned probability derivation process is, For each of the above combinations, a third individual probability is derived, which is the probability that a different delivery body was delivered. The third individual probabilities are combined with multiple combinations of the above to derive a third combined probability. The area determination process described above is: The area candidate determination process, the extraction process, and the probability derivation process are repeated. The area candidate that maximizes the third combination probability is defined as the area. Information processing device. [Claim 14] An information processing device comprising a computer that determines the delivery area for each of a plurality of delivery bodies, The aforementioned computer, Area candidate determination process that determines multiple area candidates, Based on the aforementioned area candidates, an extraction process is performed to randomly select multiple combinations of two demand points, A probability derivation process that derives the probability that the delivery body delivered to the demand point for each combination, based on past delivery data of the delivery body that delivered to the demand point. An area determination process that determines the area based on the aforementioned probability, Execute, The extraction process described above is: Within the same candidate area, multiple combinations of two demand points are randomly selected. The aforementioned probability derivation process is, For each of the aforementioned combinations, a fourth individual probability is derived, which is the probability that a different delivery body was delivered. The fourth individual probability is combined with a plurality of the above combinations to derive the fourth combined probability. The area determination process described above is: The area candidate determination process, the extraction process, and the probability derivation process are repeated. The area candidate that minimizes the fourth combination probability is defined as the area. Information processing device. [Claim 15] An information processing device comprising a computer that determines the delivery area for each of a plurality of delivery bodies, The aforementioned computer, Area candidate determination process that determines multiple area candidates, Based on the aforementioned area candidates, an extraction process is performed to randomly select multiple combinations of two demand points, A probability derivation process that derives the probability that the delivery body delivered to the demand point for each combination, based on past delivery data of the delivery body that delivered to the demand point. An area determination process that determines the area based on the aforementioned probability, Execute, The extraction process described above is: In the different candidate areas, multiple combinations of two demand points are randomly selected. The aforementioned probability derivation process is, For each combination extracted from the different area candidates, a third individual probability is derived, which is the probability that a different delivery entity delivered the package. The third individual probabilities are combined with multiple combinations of the above to derive a third combined probability. The extraction process described above is: Within the same candidate area, multiple combinations of two demand points are randomly selected. The aforementioned probability derivation process is, For each combination extracted from the same area candidate, a fourth individual probability is derived, which is the probability that a different delivery entity delivered the package. The fourth individual probability is combined with a plurality of the above combinations to derive the fourth combined probability. The area determination process described above is: The area candidate determination process, the extraction process, and the probability derivation process are repeated. The area candidate that yields the maximum result when the third combination probability is subtracted from the fourth combination probability is defined as the area. Information processing device.

Images