KR20230012801A - Optimized allocation method of logistics delivery points - Google Patents

Abstract

The present invention relates to an optimized allocation method of logistics delivery points. The allocation method of logistics delivery points, which comprises the steps of: (a) verifying a plurality of types of order information as one delivery information; (b) transmitting the delivery information to a picking unit (200) and a clustering unit (300), respectively; (c) generating clustering information classifying a plurality of delivery points; and (d) generating a dispatch signal using the clustering information.

Description

Optimized allocation method of logistics delivery points}

The present invention relates to an optimized method for allocating logistics delivery points.

Recently, as non-face-to-face orders have been activated and delivery services are provided even when ordering face-to-face, the amount of delivery is greatly increasing. To this end, it is necessary to quickly classify the delivery points and to equally allocate the delivery amount to each delivery area.

When logistics are delivered, if logistics delivery points are allocated along physical zones, there is a problem in that the distribution volumes are not evenly allocated according to a high logistics volume in one zone and a low logistics volume in another zone.

Conventionally, in order to evenly allocate the delivery amount, logistics delivery points are allocated along physical areas, but the delivery amount in areas with a large amount of delivery is artificially distributed according to circumstances.

Referring to FIG. 1 (a), a map in which delivery zones are physically distributed into reference zones 1 to 17 based on postal codes is shown. 1(b) shows a system for viewing delivery destinations in reference zones 1 to 4. Conventionally, in the system as shown in FIG. 1 (b), the delivery volume in the reference zones 1 to 4 is checked, and the delivery volume is artificially distributed even for delivery in other zones to match the delivery volume in each zone.

1(c) shows a graph in which delivery zones are allocated based on the delivery address postal code, and a graph after artificial distribution of delivery addresses is shown. In the case of assigning delivery zones based on postal codes, it can be seen that there is a large variance in delivery volume for each zone.

Also, for example, Korea Patent Publication No. 10-2016-0008422 determines the delivery order based on the delivery information of each of a plurality of delivery products associated with the delivery driver, and the delivery address of the delivery product to the delivery driver according to the delivery order. is to provide However, there was a lack of awareness about evenly distributing the delivery volume.

For another example, Korea Patent Publication No. 10-2018-0054354 relates to clustering delivery points and distributing delivery volume by clustering delivery bundle processing module 230, or according to predetermined cluster conditions such as dividing into administrative unit areas. As clustering is performed, and there is no awareness of clustering according to delivery points, there is a lack of awareness of evenly distributing delivery volume.

(Patent Document 1) Korean Patent Publication No. 10-2016-0008422

(Patent Document 2) Korean Patent Publication No. 10-2018-0054354

The present invention has been made to solve the above problems.

Specifically, the present invention is to allocate delivery points using a clustering technique, exclude points in locations where delivery is impossible, and efficiently allocate delivery points by performing a clustering technique using an optimal distance that can actually be moved.

In addition, the present invention is to even out the number of delivery points allocated per delivery manpower.

One embodiment of the present invention for solving the above problems is, (a) the input unit 100 receives order information including an order product, an order time, and a delivery point, and a plurality of orders input during a predetermined period Confirming the information as one shipping information; (b) transmitting, by the input unit 100, the delivery information to the picking unit 200 and the clustering unit 300, respectively; (c) The picking unit 200 checks the type and quantity of ordered products from a plurality of pieces of order information included in the delivery information and generates a transfer signal to be transmitted to the dispatcher 400, and the clustering unit 300 In , delivery manpower information, which is information including the number of delivery manpower, is input, and the clustering unit 300 receives optimal distance information, which is information including an optimal distance among movable routes from the navigation unit 500, and the clustering unit 300 The unit 300 generates clustering information for classifying a plurality of delivery points using the delivery information, the optimal distance information, and the delivery manpower information; and (d) the clustering unit 300 transmits the clustering information generated in step (c) to the dispatcher 400, and the dispatcher 400 uses the clustering information to determine the type of ordered product and It provides a method for allocating a delivery point, including generating a dispatch signal for distributing a quantity.

In one embodiment, the step (c) is a step using (c1) a plurality of preset center points, and the clustering unit 300 assigns the plurality of delivery points identified in the plurality of order information to each of the plurality of delivery points. forming a plurality of clusters by allocating one center point of which the distance to any one of the plurality of center points is the minimum, and forming a cluster with delivery points having the same assigned center point and the center point at that time; (c2) The clustering unit 300, for each of a plurality of clusters, moves the location of the central point included in the cluster so that the sum of the distances from the delivery points included in the cluster to the center point included in the cluster is minimized. step; (c3) The clustering unit 300, irrespective of the formed cluster, reassigns one central point at which the distance from each of the plurality of delivery points to any one of the moved central points is minimized, and the reassigned Re-forming a plurality of clusters by re-forming delivery points having the same center point and the center point at that time into one cluster; and (c4) the clustering unit 300 determines a center point that is not moved and a cluster including the center point that is not moved by repeating steps (c2) and (c3) until the center point is not moved in step (c2), , generating the clustering information including information on the determined plurality of clusters; Including, the preset number of center points may correspond to the number of delivery manpower included in the delivery manpower information.

In one embodiment, in step (c), the optimal distance information transmitted by the navigation unit 500 is a distance from a plurality of delivery points identified by the clustering unit 300 to a central point assigned to each of them. The optimal distance is included, and the step (c2) includes: (c21) checking, by the clustering unit 300, whether the central point included in the cluster is at a preset delivery impossible position; and (c22) moving, by the clustering unit 300, the longitude and latitude of the center point by predetermined values when it is confirmed in the step (c21) that the center point is in a non-deliverable position. there is.

In another embodiment of the present invention, in the step (c), (c5) the clustering unit 300 is assigned to the first to nth center points (n is a natural number of 2 or more) to the first to mth delivery points ( allocating m is a natural number equal to or greater than n) using a preset method; In the step (c5), (c51) the clustering unit 300 generates n+1th to mth center points, and the locations of the n+1th to mth center points are the first to the first to mth center points. setting a position of any one of the first to n-th center points by corresponding them one by one in the order of the n center points; (c52) checking, by the clustering unit 300, distances from the locations of the first to m-th center points to the first to m-th delivery points; (c53) The clustering unit 300 matches the first to m-th delivery points one-to-one with any one of the first to m-th center points, and the first to m-th delivery points are matched one-to-one with the first to m-th center points. matching so that the sum of the distances from the 1st to the mth central point is minimized; (c54) The clustering unit 300 moves the positions of the first to m-th center points by a preset value, and repeats steps (c52) and (c53) to move the first to m-th delivery points, respectively. confirming the sum of the distances matched one-to-one to any one of the first to m-th center points multiple times, and moving the positions of the first to m-th center points until the sum of the distances checked multiple times is minimized. ; and (c55) the clustering unit 300 determines the center point at which the first to m th center points are not moved and determines a cluster including the center point, wherein each cluster including the n+1 th center point to the m th center point is determined. It is determined to include any one of the first to n-th center points set to positions corresponding to each in step (c51) rather than including the n+1 to m-th center points, and for the determined plurality of clusters and generating the clustering information including information, wherein n corresponds to the number of delivery personnel.

In one embodiment, in step (c5), the optimal distance information transmitted by the navigation unit 500 is a distance from a plurality of delivery points checked by the clustering unit 300 to a central point assigned to each of them. including the optimal distance, and in step (c54), (c541) checking whether the clustering unit 300 has a central point at a preset delivery impossible position; and (c542) moving, by the clustering unit 300, the longitude and latitude of the center point by predetermined values if it is confirmed in the step (c541) that the center point is in a non-deliverable position.

In one embodiment, in step (d), (d1) the clustering unit 300 transmits the clustering information to the navigation unit 500; and (d2) transmitting, by the navigation unit 500, real-time dispatch information in which a delivery order between delivery points included in each cluster is determined by using the clustering information based on real-time traffic conditions to the dispatch unit 400; can include

In one embodiment, the dispatcher 400 transmits change information including a delivery schedule and a change in delivery manpower to the clustering unit 300, and the clustering unit 300 performs clustering using the change information. Step; may further include.

In one embodiment, the order information includes a user identifier, and after the step (d), the order information input to the database 600 in the step (a) and the step (c) storing the generated cluster information; and (f) learning, by the learning unit, a shipping address allocation model using the order information and the clustering information stored in the database 600 .

In one embodiment, in the step (a), the input unit 100 may further include checking a preset postal code area of the input order information.

The present invention can efficiently allocate delivery points using a clustering technique. In addition, the present invention uses a clustering technique, but moves the central point at a location where delivery is impossible, so that delivery points can be quickly clustered.

In addition, according to the present invention, the number of delivery points per delivery manpower may be evenly allocated.

In addition, since the present invention stores the delivery information and the central point at this time in the database, it is possible to increase the calculation speed when assigning the delivery point thereafter.

1 is a diagram for explaining a conventional delivery system.
2 is a schematic diagram for explaining a delivery method according to the present invention.
3 is a diagram for explaining a conventional clustering method.
4 is a diagram for explaining a clustering method according to the present invention.
5 and 6 are views for explaining a clustering method according to another embodiment of the present invention.

In some cases, in order to avoid obscuring the concept of the present invention, well-known structures and devices may be omitted or may be shown in block diagram form centering on core functions of each structure and device.

In addition, in describing the embodiments of the present invention, if it is determined that a detailed description of a known function or configuration may unnecessarily obscure the subject matter of the present invention, the detailed description will be omitted. In addition, terms to be described later are terms defined in consideration of functions in the embodiment of the present invention, which may vary according to the intention or custom of a user or operator. Therefore, the definition should be made based on the contents throughout this specification.

Referring to FIGS. 2 to 6 , a method for allocating a delivery point according to the present invention will be described.

The delivery point allocation method according to the present invention includes an input unit 100, a picking unit 200, a clustering unit 300, a dispatch unit 400, a navigation unit 500, and a database 600.

Order information is input to the input unit 100 .

The input unit 100 checks a plurality of order information input during a predetermined period as one piece of delivery information.

In this case, the preset period may be set based on a predetermined time interval. For example, the input unit 100, based on 10 am, 2 pm, and 4 pm, order information input from 4 pm to 10 am, order information input from 10 am to 2 pm, and 2 pm Order information entered from 1:00 to 4:00 p.m. may be checked as one piece of delivery information, but is not limited thereto.

The input unit 100 may check a plurality of orders entered during a certain time interval as one delivery information, but is not limited thereto, and may adjust a predetermined period according to the situation of a store or facility to which the present method is applied.

Order information includes information about the ordered product, order time and delivery point.

In this case, the information on the delivery point means to include the latitude and longitude of the delivery point.

The input unit 100 transmits delivery information to the picking unit 200 and the clustering unit 300, respectively.

In the present invention, when one piece of delivery information is confirmed, the delivery information is transmitted to the picking unit 200 and the clustering unit 300, and delivery starts using the corresponding delivery information. For example, order information input from 4:00 pm to 10:00 am is delivered first, rather than order information input from 10:00 am to 2:00 pm.

In addition, the input unit 100 may check the preset postal code area of the input order information.

In this case, the postal code area means the same area based on the postal code.

The picking unit 200 receives delivery information from the input unit 100 .

The picking unit 200 is connected to the warehouse, determines the type and number of ordered products included in the delivery information in the warehouse, and generates a transfer signal for picking the ordered products by the corresponding number and type.

Generating a transfer signal for picking ordered products in the picking unit 200 and generating clustering information in the clustering unit 300 can be performed in parallel, so that delivery information can be processed to deliver quickly. However, it goes without saying that the picking unit 200 generates the transfer signal before and after clustering information is generated by the clustering unit 300 .

In addition, the dispatcher 400 may receive clustering information from the clustering unit 300 and generate a dispatch signal for dispatching. A detailed description of this will be given later.

Delivery manpower information is input to the clustering unit 300 in advance.

Delivery personnel information includes information on the number of delivery vehicles and the number of delivery personnel. At this time, as will be described later, delivery manpower information may be updated by change information.

The clustering unit 300 receives delivery information from the input unit 100 .

The clustering unit 300 identifies a plurality of delivery points included in the delivery information, classifies the plurality of delivery points using previously input delivery manpower information, and generates clustering information.

In addition, the clustering unit 300 may generate clustering information using optimal distance information, which is an optimal distance among movable paths determined by the navigation unit 500 .

A detailed description of how the clustering unit 300 generates clustering information will be described later.

Hereinafter, a movable path determined by the navigation unit 500 means a path along which a vehicle can move on a road, and among the movable paths determined by the navigation unit 500, an optimal distance is a path in which the moving distance is the greatest among such paths. It may mean short information, but is not limited thereto, and may mean information having the shortest moving time at the current point of time determined by the navigation unit 500. Such an optimal distance may be determined in consideration of data previously accumulated in the navigation unit 500 and current road traffic conditions.

Hereinafter, a method of generating clustering information, which is an embodiment of the present invention, will be described with reference to FIGS. 3 and 4.

Hereinafter, a conventional K-means clustering technique will be described with reference to FIG. 3, and since the K-means clustering technique is an existing and well-known technique, a detailed description thereof will be omitted.

In FIG. 3, a method of clustering each delivery point when the number of delivery personnel is 3 will be described.

In FIG. 3 (a), it is shown that the center point c1, c2, and c3 are set in advance, and a plurality of delivery points are shown.

FIG. 3(b) shows that the central point is connected to the central point located at the optimal distance by determining the central point located at the optimal distance among c1, c2, and c3 at a plurality of delivery points. At this time, a delivery point having an optimal distance c1, a delivery point having an optimal distance c2, and a delivery point having an optimal distance c3 each form a cluster.

At this time, the optimal distance described in FIG. 3 does not mean the optimal distance among physically movable paths, but means the shortest distance in a straight line on the path as shown.

In FIG. 3(c), for each cluster created in FIG. 3(b), the location of the center point is moved so that the sum of the distances between c1, c2, and c3, which are the center points in the cluster, and each delivery point included in each cluster is an optimal distance. show what to do 3(d) shows that the location of the center point is moved.

When using the existing K-means clustering technique, there is a problem of not reflecting the actual path situation rather than generating a cluster using a physically movable path, so it is not actually an optimal movable distance from the center point, but it is optimal on a straight line There is a problem in that the accuracy of clustering is reduced, such as not reflecting the actual path situation because it is set as a distance, and the calculation speed also increases because the straight line distance must be recalculated every time the center point is moved.

On the other hand, in the present invention, clustering is performed, but clusters are generated using an optimal distance among actually movable paths.

Hereinafter, referring to FIG. 4 , a method of clustering a plurality of delivery points by the clustering unit 300 according to an embodiment of the present invention will be described.

The clustering unit 300 allocates a center point having a minimum distance to any one of the plurality of center points to each of the plurality of delivery points identified in the plurality of order information by using the plurality of preset center points.

At this time, the plurality of preset center points mean preset initial positions, and may be set randomly, and the number of the plurality of center points is formed to correspond to the number of delivery personnel, and then one cluster corresponds to one delivery person. It means assigned to manpower.

The clustering unit 300 forms a plurality of clusters by forming one cluster with delivery points having the same assigned center point and the center point at that time.

That is, a plurality of delivery points form a cluster with a central point closest to the location of each delivery point.

At this time, the clustering unit 300 checks the optimal distance information between the plurality of delivery points from the locations of the plurality of preset center points transmitted from the navigation unit 500, and determines the center point at which the distance is the minimum based on the optimal distance information. can do.

The clustering unit 300 moves the location of the central point included in the cluster so that the sum of the distances from the delivery points included in the cluster to the center point included in the cluster is minimized for each of the plurality of clusters.

At this time, the clustering unit 300 may check all the optimal distance information determined by the navigation unit 500 between the moved center points and the plurality of delivery points for each of the plurality of clusters.

The clustering unit 300 redistributes a central point at which the distance from each of the plurality of delivery points to any one of the moved central points is minimized, and divides the re-assigned central points into one delivery point and the same central point. By re-forming into clusters, multiple clusters are re-formed.

That is, the clustering unit 300 means checking the distance between each delivery point again based on the moved center points.

The clustering unit 300 repeatedly performs the above process until the center point does not move, determines the center point that is not moved and the cluster including the center point, and generates clustering information including information on the determined plurality of clusters.

Also, in the present invention, the clustering unit 300 may adjust the location of the center point to exclude a location where delivery is physically impossible.

The clustering unit 300 determines whether the central point is physically impossible to deliver.

Physically, the term "undeliverable location" is set in advance to mean that the location on the GPS includes mountains, rivers, seas, etc., and locations that are physically impossible to move by vehicle or on foot.

At this time, the clustering unit 300 may check from the navigation unit 500 whether or not the center point is at a delivery impossible position, but is not limited thereto.

The clustering unit 300 determines whether the central point is in a non-deliverable position, and if the central point is in a non-deliverable position, moves the longitude and latitude of the central point by a preset value.

The clustering unit 300 may repeat the above process to exclude that the center point is located at a non-deliverable location.

Accordingly, in the present invention, a route between the center point and the delivery point can be calculated using the existing navigation unit 500, and the time for generating clustering information according to the present invention can be shortened.

Hereinafter, a method of allocating a delivery address according to another embodiment of the present invention will be described with reference to FIGS. 5 and 6 .

According to this embodiment, the number of delivery points can be allocated more evenly according to the delivery manpower.

The clustering unit 300 assigns the first to m th delivery points (m is a natural number greater than or equal to n) to the preset first to n th center points (n is a natural number greater than or equal to 2) using a preset method.

In this case, the first to nth center points mean preset initial positions and may be randomly set.

The clustering unit 300 generates n+1th to mth center points, but the locations of the n+1th to mth center points correspond one by one in the order of the 1st to nth center points to any one of the 1st to nth center points. set to the position of

That is, the clustering unit 300, as will be described later, in order to match the center point and the delivery point on a one-to-one basis, the number of first to n-th center points is the same as the number of first to m-th delivery points. means to generate the mth center point.

For example, referring to FIG. 6, when m is 8, the delivery points are a, b, c, d, e, f, g h, which are the first to eighth delivery points in order, and n is 3, and the center points are A and B. , C, if the first to third center points in order, the center point is set to be equal to the number of delivery points by increasing the virtual center point to A, B, C, A, B, C, A, B.

At this time, the order of center points may be preset. For example, the order of center points of A, B, and C set in FIG. 6 may be pre-determined based on zip code area, latitude and longitude, or separately input information.

The number of center points, n, corresponds to the number of delivery personnel, and the locations of the center points are initially randomly arranged.

Accordingly, according to the present embodiment, the number of delivery points can be evenly allocated to each delivery person by adjusting the number of the center point and the number of delivery points to be the same before allocating the center point to the center point.

The clustering unit 300 checks the distances from the locations of the first to m-th center points to the first to m-th delivery points, respectively.

At this time, the clustering unit 300 checks the distance by receiving optimal distance information from the locations of the first to nth central points to the first to mth delivery points through the navigation unit 500 and checking the distance. means, but is not limited thereto.

The clustering unit 300 matches the first to m-th delivery points to one of the first to m-th center points, respectively, and the distances from the first to m-th delivery points to the one-to-one matched first to m-th center points. Matches so that the sum of is the minimum.

For example, referring to FIG. 6 , the clustering unit 300 is a center point A, B, C, A, B, C, To check the distances to A and B, it means matching a, b, c, d, e, f, g, h to any one of A, B, C, A, B, C, A, B, respectively. do. In this case, it means that one-to-one matching is performed so that the sum of distances of eight pairs of one-to-one matches is minimized.

In addition, the clustering unit 300 moves the positions of the first through m-th center points by a preset value, but repeats the above process.

When the clustering unit 300 checks the sum of the distances matched one-to-one to any one of the first to m-th center points to which the first to m-th delivery points are moved, respectively, multiple times, and the sum of the distances checked multiple times becomes minimum. The positions of the first through mth center points are moved until

The clustering unit 300 determines the central points where the first through m-th central points are not moved and determines a cluster including the central points.

In this case, the clustering unit 300 determines that each cluster including the n+1th to mth center points does not include the n+1th to mth center points, but the first to nth center points are set to corresponding positions. It is determined to include any one of the clusters, and clustering information including information on the determined plurality of clusters is generated.

According to this embodiment, it means that the n+1 to m th central points and the delivery points matched one-to-one with each are formed in the same cluster as any one of the first to n th central points set to corresponding positions.

According to this, a plurality of delivery points may be assigned to each of the first through n-th center points.

In addition, similar to the above-described embodiment, in this embodiment, when the center point of the clustering unit 300 is moved, if the center point is in a non-deliverable position, the location of the center point may be moved by a preset value. The same details as described above are omitted.

The dispatcher 400 may receive clustering information from the clustering unit 300 and generate a dispatch signal that assigns delivery points to a plurality of delivery personnel according to the clustering information. Thereafter, the delivery manpower may start delivery of the order product included in the delivery point assigned to each delivery person using the dispatch signal.

At this time, one cluster is allocated to one delivery person, and it means that delivery points included in each cluster are delivered.

In addition, when change information including a change in delivery schedule and delivery manpower occurs, the dispatcher 400 transmits the change information to the clustering unit 300, and the clustering unit 300 performs clustering using the change information. .

Accordingly, when a change in delivery schedule and a change in delivery manpower occur, the clustering unit 300 can quickly allocate a delivery point accordingly.

As described above, the navigation unit 500 may transmit the optimal distance between the center point and the delivery point to the clustering unit 300. When the clustering unit 300 performs clustering, the navigation unit 500 determines the center point and Information between delivery points is received, and an optimal distance between the central point and the delivery point may be calculated or confirmed.

As described above, the optimal distance determined by the navigation unit 500 may refer to information having the shortest moving distance among possible paths, but is not limited thereto, and the current moving time determined by the navigation unit 500 is the shortest. can mean information. Such an optimal distance may be determined in consideration of data previously accumulated in the navigation unit 500 and current road traffic conditions.

In addition, the navigation unit 500 may receive clustering information from the clustering unit 300 and determine a real-time delivery order between delivery points included in each cluster.

That is, the navigation unit 500 may determine a delivery order between delivery points by considering real-time traffic conditions of delivery points included in each cluster included in the clustering information.

Accordingly, the delivery personnel may determine the order of delivery in consideration of real-time traffic conditions.

At this time, the navigation unit 500 suffices as long as it is a system or device capable of checking or calculating an optimal distance, and is not limited to a specific device or device.

The database 600 stores order information input through the input unit 100 and clusters classified by the clustering unit 300 .

The order information may further include a user identifier.

In this case, the user identifier may include all of the user's ID, the user's business partner, the user's business type, and a plurality of delivery points of the user.

The database 600 may store information at which time and which customer orders a product by using order information.

Using the order information and clustering information stored in the database 600, a learning unit (not shown) may train a shipping destination allocation model.

In addition, the learned delivery destination allocation model may be additionally learned using order information and clustering information additionally stored in the database 600 .

When new order information is input, the delivery address allocation model may output clustering information obtained by clustering delivery points included in the new order information.

In addition, the clustering information output from the delivery destination allocation model is input to the clustering unit 300, and the center point included in the corresponding clustering information can be applied as the aforementioned preset center point. According to this, the calculation speed of the clustering unit 300 can be further increased.

In addition, since the delivery address allocation model is learned using the stored order information and clustering information, the delivery address allocation model can be used to determine the time zone at which the customer orders, the type and amount of products to be ordered, according to each region.

At this time, a method of learning order information and clusters in the learning unit is not limited to a specific method.

In the above, the present specification has been described with reference to the embodiments shown in the drawings so that those skilled in the art can easily understand and reproduce the present invention, but this is only exemplary, and those skilled in the art can make various modifications and equivalents from the embodiments of the present invention. It will be appreciated that embodiments are possible. Therefore, the scope of protection of the present invention should be defined by the claims.

100: input unit
200: picking unit
300: clustering unit
400: Distributor
500: navigation unit
600: database

Claims (9)

(a) the input unit 100 receiving order information including an order product, an order time, and a delivery point, and confirming a plurality of order information input during a predetermined period as one delivery information;
(b) transmitting, by the input unit 100, the delivery information to the picking unit 200 and the clustering unit 300, respectively;
(c) The picking unit 200 checks the type and quantity of ordered products from a plurality of pieces of order information included in the delivery information and generates a transfer signal to be transmitted to the dispatcher 400, and the clustering unit 300 In , delivery manpower information, which is information including the number of delivery manpower, is input, and the clustering unit 300 receives optimal distance information, which is information including an optimal distance among movable routes from the navigation unit 500, and the clustering unit 300 The unit 300 generates clustering information for classifying a plurality of delivery points using the delivery information, the optimal distance information, and the delivery manpower information; and
(d) The clustering unit 300 transmits the clustering information generated in step (c) to the dispatcher 400, and the dispatcher 400 uses the clustering information to determine the type and quantity of ordered products. Generating a dispatch signal for dispatching; including,
How to assign a delivery point.
According to claim 1,
In step (c),
(c1) a step of using a plurality of preset center points, wherein the clustering unit 300 has a minimum distance from each of the plurality of delivery points identified in the plurality of order information to any one of the plurality of center points forming a plurality of clusters by allocating one central point, and forming a cluster with delivery points having the same assigned central point and the central point at that time;
(c2) The clustering unit 300, for each of a plurality of clusters, moves the location of the central point included in the cluster so that the sum of the distances from the delivery points included in the cluster to the center point included in the cluster is minimized. step;
(c3) The clustering unit 300, irrespective of the formed cluster, reassigns one central point at which the distance from each of the plurality of delivery points to any one of the moved central points is minimized, and the reassigned Re-forming a plurality of clusters by re-forming delivery points having the same center point and the center point at that time into one cluster; and
(c4) The clustering unit 300 determines the center point that is not moved and the cluster including it by repeating steps (c2) and (c3) until the center point is not moved in step (c2), generating the clustering information including information on the determined plurality of clusters; including,
The preset number of center points corresponds to the number of delivery manpower included in the delivery manpower information,
How to assign a delivery point.
According to claim 2,
In step (c),
The optimal distance information transmitted by the navigation unit 500 includes an optimal distance that is a distance from a plurality of delivery points identified by the clustering unit 300 to a central point assigned to each,
In step (c2),
(c21) checking, by the clustering unit 300, whether or not the central point of the cluster is at a preset delivery impossible location; and
(c22) the clustering unit 300, if it is confirmed in the step (c21) that the center point is in a non-deliverable position, moving the longitude and latitude of the center point by a predetermined value; further comprising,
How to assign a delivery point.
According to claim 1,
In step (c),
(c5) allocating, by the clustering unit 300, first to m-th delivery points (m is a natural number equal to or greater than n) to preset first to n-th central points (n is a natural number greater than or equal to 2) using a preset method; ; including,
In step (c5),
(c51) The clustering unit 300 generates n+1th to mth center points, and the locations of the n+1th to mth center points correspond to each other in order of the first to nth center points, so that the first to nth center points correspond to each other. to set to any one of the n-th central points;
(c52) checking, by the clustering unit 300, distances from the locations of the first to m-th center points to the first to m-th delivery points;
(c53) The clustering unit 300 matches the first to m-th delivery points one-to-one with any one of the first to m-th center points, and the first to m-th delivery points are matched one-to-one with the first to m-th center points. matching so that the sum of the distances from the 1st to the mth central point is minimized;
(c54) The clustering unit 300 moves the positions of the first to m-th center points by a preset value, and repeats steps (c52) and (c53) to move the first to m-th delivery points, respectively. confirming the sum of the distances matched one-to-one to any one of the first to m-th center points multiple times, and moving the positions of the first to m-th center points until the sum of the distances checked multiple times is minimized. ; and
(c55) The clustering unit 300 determines the center point at which the first to m th center points are not moved and determines a cluster including the center point, each cluster including the n+1 th center point to the m th center point It is determined to include any one of the first to n-th center points set to corresponding positions in step (c51) rather than including the n+1 to m-th center points, and information on the determined plurality of clusters. Generating the clustering information including; Including,
n corresponds to the number of delivery personnel,
How to assign a delivery point.
According to claim 4,
In step (c5),
The optimal distance information transmitted by the navigation unit 500 includes an optimal distance that is a distance from a plurality of delivery points identified by the clustering unit 300 to a central point assigned to each,
In the step (c54),
(c541) checking, by the clustering unit 300, whether or not the center point is at a preset delivery impossible position; and
(c542) the clustering unit 300 moving the longitude and latitude of the center point by predetermined values when it is determined in step (c541) that the center point is in a non-deliverable position;
How to assign a delivery point.
According to claim 1,
In step (d),
(d1) transmitting, by the clustering unit 300, the clustering information to the navigation unit 500; and
(d2) transmitting, by the navigation unit 500, real-time dispatch information determined by determining a delivery order between delivery points included in each cluster by using the clustering information based on real-time traffic conditions to the dispatch unit 400; doing,
How to assign a delivery point.
According to claim 1,
The dispatcher 400 transmits change information including a delivery schedule and a change in delivery manpower to the clustering unit 300, and the clustering unit 300 performs clustering using the change information; further comprising ,
How to assign a delivery point.
According to any one of claims 1 to 5,
The order information includes a user identifier,
After step (d),
(e) storing the order information input in step (a) and the cluster information generated in step (c) in a database 600; and
(f) learning, by the learning unit, a shipping address allocation model using the order information and the clustering information stored in the database 600;
How to assign a delivery point.
According to claim 1,
In step (a),
The input unit 100 confirms a preset postal code area of the input order information; further comprising,
How to assign a delivery point.

Images