WO2018227859A1 - 配送区域划分方法、装置、电子设备及计算机可读存储介质 - Google Patents

配送区域划分方法、装置、电子设备及计算机可读存储介质 Download PDF

Info

Publication number
WO2018227859A1
WO2018227859A1 PCT/CN2017/109999 CN2017109999W WO2018227859A1 WO 2018227859 A1 WO2018227859 A1 WO 2018227859A1 CN 2017109999 W CN2017109999 W CN 2017109999W WO 2018227859 A1 WO2018227859 A1 WO 2018227859A1
Authority
WO
WIPO (PCT)
Prior art keywords
cluster
clusters
order
delivery
area corresponding
Prior art date
Application number
PCT/CN2017/109999
Other languages
English (en)
French (fr)
Inventor
杨秋源
徐明泉
黄绍建
咸珂
陈进清
饶佳佳
Original Assignee
北京小度信息科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 北京小度信息科技有限公司 filed Critical 北京小度信息科技有限公司
Publication of WO2018227859A1 publication Critical patent/WO2018227859A1/zh
Priority to US16/713,318 priority Critical patent/US20200118075A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G06Q10/043—Optimisation of two dimensional placement, e.g. cutting of clothes or wood
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/083—Shipping
    • G06Q10/0838—Historical data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/29—Geographical information databases
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/083—Shipping
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0633—Lists, e.g. purchase orders, compilation or processing
    • G06Q30/0635—Processing of requisition or of purchase orders

Definitions

  • the present disclosure relates to the field of Internet technologies, and in particular, to a method, an apparatus, an electronic device, and a computer readable storage medium for distributing a distribution area.
  • O2O Online To Offline
  • the scheduling of orders by the logistics dispatching system is often based on the distribution area (also referred to as the business circle in some scenarios).
  • the distribution area also referred to as the business circle in some scenarios.
  • an order generated in a delivery area is assigned to a delivery person belonging to the delivery area for distribution.
  • the division of the distribution area of the city is mainly based on the knowledge of the city’s road traffic conditions, the distribution of merchants, and the distribution of users. Distribution areas where the areas do not intersect. Therefore, when scheduling an order for the city, it is scheduled based on a plurality of distribution areas obtained by manual division.
  • the distribution area delineated by manual experience may be unreasonable, which may result in waste of delivery capacity when order is scheduled.
  • a distribution area based on manual experience may generate a large number of orders across distribution areas.
  • these cross-distribution regional orders will lead to the return of the delivery personnel, which is a waste of delivery capacity.
  • the delivery scenario of the sale order may include the location information of the pickup location and the delivery location, and the order is when the pickup location and the delivery location of an order correspond to different delivery areas. This is an order across distribution areas.
  • the above-mentioned order-based scheduling mechanism based on the delivery area is simply that the delivery personnel can only accept orders for the delivery area to which they belong.
  • the delivery area corresponding to the order may be determined by determining the order of the delivery area according to the delivery area corresponding to the pickup location of the order. Then, when scheduling the order, the order needs to be assigned to a delivery person belonging to the delivery area A to complete the delivery of the order.
  • the distribution area to which the delivery personnel belong can be determined according to the registration information provided by the delivery personnel in the logistics dispatch system, that is, the delivery personnel are registered to a delivery area when registering. Based on this, assuming that a delivery person belongs to the delivery area A, it can only accept the order of the delivery area A, meaning that it can only accept the order of the pickup location in the delivery area A.
  • the above-mentioned empty returning phenomenon can be understood as follows: It is assumed that the pickup position of an order across the delivery area corresponds to the delivery area A, and the delivery position corresponds to the delivery area B. At this time, the order is determined to be the order of the delivery area A. It will be assigned to a delivery person in the delivery area A for delivery. After the delivery personnel completes the delivery to the delivery location in the delivery area B, it is required to return to the home delivery area A to continue accepting the order of the delivery area A, and cannot accept the order in the delivery area B within the delivery area B. Because the delivery person belongs to the delivery area A, it can only accept the order of the delivery area A to which it belongs. Thus, the order across the distribution area causes an empty run phenomenon of the delivery personnel in the process of returning to the delivery area A, wasting the delivery capacity.
  • the embodiments of the present disclosure provide a method, an apparatus, an electronic device, and a computer readable storage medium for allocating a distribution area, so as to improve the rationality of the distribution area range division result.
  • an embodiment of the present disclosure provides a method for dividing a delivery area, including:
  • an embodiment of the present disclosure provides a distribution area dividing apparatus, including:
  • a clustering module configured to cluster the plurality of historical orders according to geographical similarity between the plurality of historical orders, to obtain N clusters, N ⁇ 1;
  • An obtaining module configured to acquire, according to a geographic location of a historical order included in each of the N clusters, a coverage area corresponding to each of the N clusters;
  • a first determining module configured to determine, according to a coverage area corresponding to each of the N clusters, a delivery area corresponding to each of the N clusters.
  • the structure of the above-mentioned distribution area dividing apparatus includes a processor and a memory for storing a program supporting the distribution area dividing means to execute the distribution area dividing method in the above first aspect, the processor being Configured to execute a program stored in the memory.
  • the distribution area dividing means may further comprise a communication interface for the distribution area dividing means to communicate with other devices or communication networks.
  • an embodiment of the present disclosure provides a computer readable storage medium for storing computer software instructions for a distribution area dividing apparatus, including a program for performing the distribution area dividing method in the first aspect.
  • the distribution area dividing method, device, electronic device and computer readable storage medium acquire a large number of actual historical orders generated, and combine these historical orders by combining the geographical similarities between the historical orders.
  • Clustering process clustering historical orders with similar geographical locations into a cluster, thereby obtaining N clusters.
  • a closed coverage area including the geographic locations of all historical orders in the cluster may be delineated to cover the area Determine the scope of the distribution area corresponding to the cluster.
  • the clustering process can make different clusters
  • the coverage area is relatively independent, so that the automatic demarcation of the delivery area is realized based on the clustering processing of a large number of historical orders.
  • the present solution can also be understood as a scheme for continuously optimizing the range of each delivery area. It is assumed that each delivery area that has been in existence before the execution of the present scheme is referred to as an original delivery area. For the above historical order, based on the positioning of the geographical location of each historical order in the original delivery area, it is assumed that some of the historical orders belong to the order across the delivery area.
  • the historical orders that are originally across the distribution area are clustered into a certain cluster based on the geographical similarity with other historical orders, and the coverage area of the cluster will be The geographic location of the historical order across the distribution area is included, so that the scope of the distribution area delineated according to the coverage area of such clusters will cover the geographical location of the historical order across the distribution area, and then the historical order of the cross-distribution area will appear again When the location corresponds to a new order, the new order will no longer be an order across the distribution area. Therefore, according to the scheme, the continuous optimization and division of the distribution area can effectively reduce the generation of orders across the distribution area.
  • FIG. 1 is a flowchart of Embodiment 1 of a method for dividing a delivery area according to an embodiment of the present disclosure
  • step 102 is a flow chart of an alternative implementation of step 102 in the embodiment shown in FIG. 1;
  • step 3a is a flow chart of an alternative implementation of step 103 in the embodiment shown in FIG. 1;
  • FIG. 3b-3f are schematic views of processes when the embodiment shown in Fig. 3a is performed;
  • FIG. 4 is a flowchart of Embodiment 2 of a method for dividing a delivery area according to an embodiment of the present disclosure
  • FIG. 5 is a schematic structural diagram of Embodiment 1 of a distribution area dividing apparatus according to an embodiment of the present disclosure
  • FIG. 6 is a schematic structural diagram of Embodiment 2 of a distribution area dividing apparatus according to an embodiment of the present disclosure
  • FIG. 7 is a schematic structural diagram of Embodiment 3 of a distribution area dividing apparatus according to an embodiment of the present disclosure.
  • FIG. 8 is a schematic structural diagram of Embodiment 4 of a distribution area dividing apparatus according to an embodiment of the present disclosure
  • FIG. 9 is a schematic structural diagram of an electronic device corresponding to a delivery area dividing apparatus according to an embodiment of the present disclosure.
  • first, second, third, etc. may be used to describe XXX in embodiments of the present disclosure, these XXX should not be limited to these terms. These terms are only used to distinguish XXX.
  • the first XXX may also be referred to as the second XXX without departing from the scope of the embodiments of the present disclosure.
  • the second XXX may also be referred to as the first XXX.
  • the words “if” and “if” as used herein may be interpreted to mean “when” or “when” or “in response to determining” or “in response to detecting.”
  • the phrase “if determined” or “if detected (conditions or events stated)” may be interpreted as “when determined” or “in response to determination” or “when detected (stated condition or event) “Time” or “in response to a test (condition or event stated)”.
  • FIG. 1 is a flowchart of Embodiment 1 of a method for dividing a delivery area according to an embodiment of the present disclosure.
  • the method for dividing a delivery area provided by this embodiment may be performed by a distribution area division apparatus, and the distribution area division apparatus may be implemented as software. Or implemented as a combination of software and hardware, the distribution area dividing device can be integrated in a device on the side of the logistics scheduling platform, such as a server.
  • the method includes the following steps:
  • the plurality of historical orders may be obtained by collecting orders corresponding to one or more original delivery areas of a certain city within a certain historical time.
  • the original delivery area refers to a delivery area that has been drawn before performing the delivery area division method provided by the embodiment of the present disclosure.
  • the distribution area division method provided by the embodiment of the present disclosure can also be regarded as an optimization method of the distribution area division range.
  • the belonging distribution area selects all or part of the historical orders corresponding to the one or more original delivery areas as the plurality of historical orders in all the historical orders received in a certain historical time.
  • the geographic location corresponding to the order may include a pickup location and a delivery location.
  • the pickup location often corresponds to the address of a merchant
  • the delivery location often corresponds to The address of the user who purchased the item.
  • the geographic similarity between the historical orders on which the plurality of historical orders are clustered may be measured by at least one of the following parameters: the distance between the pickup locations, and the delivery The distance between the locations, the distance between the pickup location and the center of the location of the shipping location.
  • clustering multiple historical orders may be implemented by setting a similarity threshold in advance, and calculating a geographic similarity between any two historical orders, so that the geographic similarity of each other is greater than or Historical orders equal to the similarity threshold are clustered into the same cluster.
  • a historical order may be randomly selected from a plurality of historical orders as a reference order, and the reference order leads to a cluster, and then, based on the remaining un-clustered historical orders, respectively
  • the geographic similarity between the reference orders, and the historical order in which the geographic similarity between the reference orders and the reference order is greater than the similarity threshold is selected from the remaining un-clustered historical orders, and is added to the cluster corresponding to the reference order, thereby A cluster corresponding to the base order is formed.
  • an unclustered historical order refers to a historical order that has not been clustered into a cluster. Then, for the remaining un-clustered historical orders, the selection of the above-mentioned reference orders and the processing of the clusters corresponding to the reference orders are performed until all the historical orders are clustered into a certain cluster.
  • the plurality of historical orders include order 1, order 2, order 3, order 4, order 5, and order 6.
  • the first selected benchmark order is order 1, which leads to cluster 1, according to other un-clustering historical orders, namely, order 2, order 3, order 4, order 5, and order 6 respectively. Comparing the degree thresholds, assuming that the geographical similarity between the order 2 and the order 1 and the order 1 is greater than the similarity threshold, the order 2 and the order 3 are clustered into the cluster 1 corresponding to the order 1, and thus, in the cluster 1 Includes order 1, order 2, order 3.
  • the second selected base order is order 6, and the cluster 2 is drawn.
  • the order 4 and the order 5 are assumed. If the geographic similarity of 6 is greater than the similarity threshold, the order 4 is clustered into the cluster 2 corresponding to the order 6, so that the cluster 2 includes the order 4 and the order 6. After that, only order 5 remains, and order 5 forms cluster 3.
  • a historical order can be randomly selected from a plurality of historical orders as a reference order to bring up a cluster.
  • the un-cluster order may be clustered according to the geographic similarity between the un-cluster order in the plurality of historical orders and the historical order in the cluster corresponding to the reference order.
  • the clustering process is: each time a historical order is added to the cluster, the next joined historical order needs to satisfy the condition: the geography between the historical orders that have been added to the cluster.
  • the position similarity is greater than or equal to the similarity threshold.
  • the plurality of historical orders include order 1, order 2, order 3, order 4, order 5, and order 6.
  • the first selected base order is order 1, which leads to cluster 1.
  • the unclustered historical order at this time is Order 2, Order 3, Order 4, Order 5, and Order 6, assuming that the order 2 is randomly selected to determine whether it can enter the cluster 1 , assuming the geography of the order 2 and the order 1
  • the location similarity is greater than the similarity threshold and is added to cluster 1.
  • the remaining un-clustering history orders are updated to order 3, order 4, order 5, and order 6, and further, an order 1 and order 2 that have been added to the cluster 1 are selected from the un-clustered historical order.
  • the order 5 is added to the cluster 1.
  • the remaining un-clustered historical orders at this time that is, the historical orders with the similarity degree between the order 3, the order 4, and the order 6 that have not been added to the historical orders in the joined cluster 1 are greater than the similarity threshold.
  • the resulting cluster 1 includes order 1, order 2, and order 5. Thereafter, the above clustering process is performed for the remaining historical orders until all historical orders are clustered into a certain cluster.
  • the above Geographic similarity is generally expressed by weighted results of multiple metrics.
  • the weighting coefficient can be set to 1, and the weighting coefficient of each metric parameter can also be empirically set.
  • the geographical similarity between the two historical orders needs to consider the distance between the pickup addresses and the distance between the delivery addresses.
  • Setting a corresponding similarity threshold for each metric parameter such as setting a threshold 1 corresponding to the distance between the pickup addresses, and a threshold 2 corresponding to the distance between the delivery addresses, at this time, the geographical similarity of the two historical orders
  • the similarity score 1 is taken as an example, and the correspondence between the similarity score 1 and the corresponding pickup address has a preset correspondence relationship, for example, the correspondence may be set by a preset function, that is, The function mapping relationship between the distance between the pickup addresses and the similarity score may be preset: f(L), where L represents the distance between the pickup addresses, and f represents the function mapping relationship.
  • the historical orders are clustered, and it is assumed that N clusters can be obtained, wherein the geographic location is further based on the clustering processing.
  • Similar historical orders are clustered into a cluster, that is, historical orders whose geographic similarity is greater than or equal to the preset similarity threshold are clustered into a cluster.
  • a certain historical order among the plurality of historical orders is an order across the delivery area, that is, the pickup location and the delivery position of the historical order are respectively assumed to correspond to different original delivery areas.
  • the historical order across the delivery area may be clustered into a certain cluster based on the geographical similarity with other historical orders, because there may be a certain number of historical orders.
  • Historical orders like this historical order, are orders across distribution areas and the delivery areas spanned are the same.
  • the distribution area division method provided by the embodiment of the present disclosure can also be regarded as an optimization method of the distribution area division range, because if there are more cross-distribution area orders between the two original delivery areas, this may be The division of the two original delivery areas is not appropriate. If a cluster contains a historical order across the original distribution area, the coverage area of the cluster will contain the pick-up address and shipping address of the historical order across the original delivery area, thereby performing the cluster. When the distribution area is divided, the newly-divided delivery area will cover the pick-up address and delivery address of the historical order across the original delivery area, so that the historical order across the original delivery area will not be in the newly-divided delivery area. Then for the orders across the distribution area, the newly allocated delivery area is likely to greatly reduce the cross Delivery area orders are generated.
  • the multiple historical orders are divided into different clusters. Among them, a historical order will only be clustered into a cluster. Further, for each cluster, optionally, a coverage area corresponding to the cluster may be demarcated on the preset map according to a geographic location corresponding to the historical order included in the cluster, the coverage area The geographic location corresponding to all historical orders in the cluster is within the coverage area.
  • the preset map may be a basic map for performing the distribution area division of the above-mentioned city, and includes a road, a preset name of some delivery areas, a name of the merchant, and the like, optionally, the pre- It is also possible to display an identifier of the original delivery area range corresponding to each delivery area name on the map.
  • the identifier of the historical order included in any cluster may be located on the preset map, and then, on the preset map, The coverage area of any cluster.
  • the identifier of the historical order may be a pattern of a certain shape, such as a dot, a square, a small red flag, used to identify a pickup address and a delivery address of the corresponding historical order, and therefore, each historical order may have two Identification, one for identifying the pick-up address and one for identifying the shipping address.
  • the determining of the coverage area corresponding to any type of cluster may be obtained by using a convex hull algorithm, that is, obtaining the coverage area by obtaining a minimum convex polygon covering a geographic location corresponding to all historical orders in the cluster.
  • a Cartesian coordinate system may be established in advance on the preset map.
  • each historical order identifier may be obtained. The corresponding coordinate position in this coordinate system.
  • the commonly used convex hull algorithm such as Graham scanning method and Jarvis stepping method can be used to solve the coverage area corresponding to each cluster.
  • the number of history orders is at least a certain number of thresholds, and/or the number of days collected has reached at least a certain number of days threshold.
  • the distribution area corresponding to the cluster of any one of the clusters is determined to be its corresponding coverage area.
  • the distribution area corresponding to the cluster of any one of the clusters may be determined as a union of the corresponding coverage area and the original delivery area corresponding to the cluster.
  • the first distribution area corresponding to the cluster is first determined. Specifically, because the historical order that is clustered into a cluster is basically composed of historical orders corresponding to the same original delivery area based on the foregoing clustering processing based on the geographical similarity, the original distribution corresponding to a certain cluster The area can be determined based on the original shipping area corresponding to the vast majority of historical orders in the cluster.
  • the category The coverage area of the cluster will include the geographic location of the historical order across the distribution area, that is, the location of the pickup and the delivery location, so that the range of the delivery area delineated according to the coverage area of the cluster will also cover the cross-distribution area.
  • the geographic location of the historical order then when a new order corresponding to the geographic location of the historical order across the delivery area appears, the new order will no longer be an order across the delivery area. Therefore, the continuous optimization division of the delivery area based on the distribution area division method provided by the embodiment can effectively reduce the generation of orders across the distribution area.
  • the large number of historical orders are clustered, and the distribution area of the corresponding clusters is determined based on the coverage area corresponding to each cluster, and the delivery area is realized.
  • the automatic delineation, and can make the allocated delivery area can be beneficial to reduce the generation of orders across the distribution area, thereby facilitating the efficient use of the delivery capacity.
  • FIG. 2 is a flowchart of an optional implementation manner of step 102 in the embodiment shown in FIG. 1. As shown in FIG. 2, the following steps may be included:
  • each of the N clusters obtains the corresponding order line segments of the N clusters, and the order line segment corresponds to the historical order one by one, and each order line segment takes the picking position of the corresponding historical order and
  • the shipping location is the endpoint.
  • the identifier of the historical order included in the cluster may be drawn on the preset map. Further, for any type of cluster, referring to the description in the foregoing embodiment, the coverage area can be obtained by using an algorithm such as a convex hull algorithm to obtain a minimum closed polygon covering all the order line segments corresponding to the cluster.
  • the identifier of the historical order may be in the form of an order line segment.
  • a historical order corresponds to an order line segment, and the two endpoints of the order line segment are the picking position and the delivery position of the historical order respectively.
  • a historical order may be randomly selected from a plurality of historical orders, and a pick-up address or a shipping address of the historical order is used as a reference address to determine a pick-up address and a shipping address of each historical order relative to the reference address. The relative position, which in turn connects the pick-up address and shipping address of each historical order, to the order line segment for each historical order.
  • the coverage areas obtained may not be completely independent coverage areas when possible, that is, there may be overlapping of partial coverage areas.
  • FIG. 3a is a flowchart of an optional implementation of step 103 in the embodiment shown in FIG. 1. As shown in FIG. 3a, the following steps may be included:
  • the coverage areas of the two clusters are considered to overlap: if there is at least one order line segment, the order line segment corresponds to The picking location is in a cluster of the two clusters, and the delivery location corresponding to the order segment is in another cluster of the two clusters.
  • the distribution area corresponding to any of the clusters Ni It can be directly determined as the coverage area corresponding to any of the clusters Ni.
  • the attribution of the overlapping region may be determined. Adjusting the coverage area of the cluster Ni also means that the coverage area of the cluster Nj is also adaptively adjusted.
  • the determination of the attribution of the overlapping area may be determined according to the number of order line segments in which the cluster cluster Ni and the cluster cluster Nj fall within the overlapping area, respectively.
  • the order line may be considered as entering the overlapping area, and the order line segment is not necessarily required to be completely overlapped. It is considered that the interior of the area falls into the overlapping area.
  • the historical order is clustered into the cluster Ni or the cluster Nj after being clustered. Therefore, according to whether the historical order corresponding to the order line segment is clustered into the cluster Ni or the cluster Nj, the cluster corresponding to the order segment can be determined. That is to say, in the overlapping area, the order line segment corresponding to the cluster Ni is falling into the overlapping area, and the corresponding historical order is gathered into the order line segment in the cluster Ni.
  • the number of order line segments corresponding to the cluster Ni is larger than the number of order line segments corresponding to the cluster Nj, and the overlapping area can be considered to belong to the cluster Ni, and at this time, the coverage of the cluster Ni The area will expand, and the coverage area of the cluster Nj may become smaller due to the need to remove overlapping areas.
  • the number of order line segments corresponding to the cluster Ni is smaller than the number of order segments corresponding to the cluster Nj, and the overlapping area can be considered to belong to the cluster Nj.
  • the coverage area of Ni may become smaller due to the need to remove the overlapping area, and the coverage area of the cluster Nj may expand.
  • the coverage area of the cluster Ni is enlarged, and the coverage area of the cluster Nj is reduced as an example to illustrate an optional determination manner of the coverage area of the cluster Ni and the cluster Nj at this time:
  • the attribution to the cluster Ni is equivalent to the fact that the historical orders corresponding to the order segments in the overlapping area are assigned to the cluster Ni.
  • the historical orders contained in the cluster Ni and the cluster Nj are updated respectively:
  • the class cluster Ni will be added to the historical order corresponding to the order line segment in the overlapping area.
  • the class order Nj will be removed from the historical order corresponding to the order line segment in the overlapping area, and thus, according to the cluster Ni and the class The geographic location of the updated historical order in the cluster Nj, adjusting the coverage area of the cluster Ni and the cluster Nj.
  • the determination of the attribution of the above overlapping area is to determine the attribution of the overlapping area as a whole.
  • the distribution of the order segments may exhibit such characteristics as being closer to the cluster Ni.
  • the class cluster Ni corresponds to more order line segments
  • the class cluster Nj corresponds to more order line segments.
  • closer to the cluster Ni means that the distance of the coverage boundary of the relative cluster Ni is greater than the coverage boundary of the relative cluster Nj. the distance.
  • a smaller-granularity attribution determination process may be performed on the overlapping area.
  • the overlapping area may be determined as a whole, or the overlapping area may be subdivided into a plurality of sub-areas, and the attribution of the plurality of sub-areas may be determined.
  • the attribution of the overlapping area may be determined as a whole, or the overlapping area may be subdivided into a plurality of sub-areas, and the attribution of the plurality of sub-areas may be determined. The attribution of the overlapping area.
  • the overlapping region may be first meshed to obtain M meshes, as shown in FIG. 3d.
  • the shape and size of the grid can be set according to actual needs.
  • the meshing of the overlapping area may be implemented by directly meshing the overlapping area, or may be implemented indirectly by meshing the entire coverage area of all N clusters. of.
  • the number of order line segments corresponding to the clusters Ni and the clusters Nj in each grid is determined to be respectively in each grid according to the clusters Ni and the clusters Nj. The number of corresponding order line segments determines the clusters to which the M grids belong.
  • the order line segment may fall into multiple grids, that is, intersect with multiple grids.
  • an order line segment corresponding to the cluster cluster Ni intersects with grid 1 and grid 2, then, When the cluster Ni is determined by the number of corresponding order line segments in the grid 1, the order line segment is calculated as the cluster Ni in a corresponding order line segment in the grid 1, and the cluster Ni is in the grid 2 When the number of corresponding order line segments is determined, the order line segment is also calculated as the number of the corresponding order line segments in the grid 2 as the cluster Ni.
  • a phenomenon may occur: for one of the grids, the cluster to which the grid belongs Unlike the clusters to which other meshes belong, the attribution of the mesh often needs to be corrected.
  • the cluster to which the mesh Mk belongs may be updated according to the cluster to which the plurality of adjacent meshes of the mesh Mk belong.
  • the grid displayed by the difference originally belongs to the cluster Nj, and most of the grids belong to the cluster Ni, especially the grid closer to the cluster Nj than the cluster.
  • the cluster Ni indicates that the grid is an anomalous grid, and the correction of the belonging cluster is required to be corrected to the cluster Ni, as shown in Fig. 3e.
  • the reason for this phenomenon may be that the order line segments corresponding to the cluster Nj in the overlapping area are exactly distributed in the coverage of a certain grid, and the order line segments corresponding to the cluster cluster Ni are widely distributed in the grid around the grid. distributed.
  • the coverage area corresponding to the cluster Ni can be adjusted according to the clusters to which the M grids belong. Specifically, the mesh corresponding to the cluster Ni in the overlapping region is divided into the coverage region of the cluster Ni, and the mesh corresponding to the cluster Nj in the overlapping region is classified into the coverage region of the cluster Nj, as shown in FIG. 3f. Shown.
  • the distribution area corresponding to the cluster Ni is a coverage area corresponding to the adjusted cluster Ni.
  • the determination of the belonging clusters can achieve reasonable adjustment of various cluster coverage areas, so that a more reasonable distribution area is determined based on the adjusted coverage areas of the various clusters.
  • FIG. 4 is a flowchart of Embodiment 2 of a method for dividing a delivery area according to an embodiment of the present disclosure. As shown in FIG. 4 , based on the foregoing embodiments, the embodiment shown in FIG. 1 is taken as an example, and after step 103, Including the following steps:
  • whether the delivery location and the delivery location are located in the same original delivery area may be determined according to whether the order is an inter-distribution area, thereby, in the original delivery In the area, the ratio of the original cross-delivery area order corresponding to the plurality of historical orders may be determined according to the ratio of the number of orders across the distribution area to the total number of the plurality of historical orders.
  • the orders of the delivery areas can be arranged according to the plurality of historical orders.
  • the ratio of the number to the total number of the plurality of historical orders determines a new cross-delivery area order ratio corresponding to the plurality of historical orders. If the original cross-distribution area order ratio is greater than the new cross-delivery area order ratio, the newly identified N categories are indicated.
  • the distribution area corresponding to the cluster is reasonable, and conversely, it indicates that the newly determined distribution area corresponding to the N clusters is unreasonable. If it is unreasonable, the original delivery area can be kept unchanged, that is, the original delivery area is not currently adjusted.
  • the delivery area dividing device of one or more embodiments of the present disclosure will be described in detail below. Those skilled in the art will appreciate that these distribution area dividing devices can be constructed using commercially available hardware components configured by the steps taught by the present solution.
  • FIG. 5 is a schematic structural diagram of Embodiment 1 of a distribution area dividing apparatus according to an embodiment of the present disclosure. As shown in FIG. 5, the apparatus includes: a clustering module 11, an obtaining module 12, and a first determining module 13.
  • the clustering module 11 is configured to perform clustering processing on the plurality of historical orders according to geographical similarity between the plurality of historical orders to obtain N clusters, N ⁇ 1.
  • the obtaining module 12 is configured to acquire, according to the geographic location of the historical order included in each of the N clusters, a coverage area corresponding to each of the N clusters.
  • the first determining module 13 is configured to determine, according to a coverage area corresponding to each of the N clusters, a delivery area corresponding to each of the N clusters.
  • the geographic location includes a pickup location and a delivery location; the geographic similarity is measured by at least one of the following parameters: a distance between the pickup locations, a distance between the delivery locations, a pickup location, and a delivery The distance between the center point of the location of the cargo location.
  • the apparatus shown in FIG. 5 can perform the method of the embodiment shown in FIG. 1.
  • FIG. 6 is a schematic structural diagram of Embodiment 2 of a distribution area division apparatus according to an embodiment of the present disclosure.
  • the acquisition module 12 includes: a first acquisition unit 121, Two acquisition unit 122.
  • the first obtaining unit 121 is configured to acquire, according to the geographic location of the historical order included in each of the N clusters, an order line segment corresponding to each of the N clusters, and the order line segment corresponds to the historical order one by one, each order
  • the single line segment is the end point of the pickup position and delivery position of the corresponding historical order.
  • a second acquiring unit 122 configured to acquire a minimum closed polygon corresponding to each of the N clusters, wherein the minimum closed polygon causes the order line segments belonging to the corresponding cluster to be included in the minimum closed polygon, the N The smallest closed polygon corresponding to each of the clusters is used as a coverage area corresponding to each of the N clusters.
  • the apparatus shown in FIG. 6 can perform the method of the embodiment shown in FIG. 2.
  • FIG. 7 is a schematic structural diagram of Embodiment 3 of a distribution area dividing apparatus according to an embodiment of the present disclosure.
  • the first determining module 13 includes: a first determining unit 131 and a second.
  • the first determining unit 131 is configured to: for any one of the N clusters, if the coverage area corresponding to the cluster Ni does not overlap with the coverage area corresponding to the other cluster Nj, determine the class
  • the third determining unit 133 is configured to determine a cluster of the overlapping area according to the number of the order line segments.
  • the adjusting unit 134 is configured to adjust a coverage area corresponding to the cluster Ni according to the cluster to which the overlapping area belongs.
  • the fourth determining unit 135 is configured to determine that the delivery area corresponding to the cluster Ni is the coverage area corresponding to the adjusted cluster Ni.
  • the second determining unit 132 is specifically configured to: perform mesh division on the overlapping area to obtain M grids; determine the cluster Ni and the cluster Nj in the M grids The number of order line segments corresponding to each grid.
  • the third determining unit 133 is specifically configured to: determine the M according to the number of order line segments respectively corresponding to the clusters Ni and the clusters Nj in each grid of the M grids The clusters to which each grid belongs.
  • the adjusting unit 134 is specifically configured to: adjust a coverage area corresponding to the cluster Ni according to a cluster to which the M grids belong.
  • the first determining module 13 further includes: an identifying unit 136 and an updating unit 137.
  • the identifying unit 136 is configured to: for any one of the M grids, according to a position of the mesh Mk in the overlapping area and a plurality of adjacent meshes of the mesh Mk The cluster of clusters identifies whether the grid Mk is an anomalous grid.
  • the updating unit 137 is configured to update the cluster to which the mesh Mk belongs according to the cluster to which the plurality of adjacent mesh belongs if the mesh Mk is an abnormal mesh.
  • the apparatus shown in FIG. 7 can perform the method of the embodiment shown in FIG. 3a.
  • the apparatus shown in FIG. 7 can perform the method of the embodiment shown in FIG. 3a.
  • the parts not described in detail in this embodiment reference may be made to the related description of the embodiment shown in FIG. 3a.
  • the implementation process and technical effects of the technical solution refer to the description in the embodiment shown in FIG. 3a, and details are not described herein again.
  • FIG. 8 is a schematic structural diagram of Embodiment 4 of a distribution area dividing apparatus according to an embodiment of the present disclosure. As shown in FIG. 8, the apparatus further includes: a second determining module 21, a third determining, as shown in FIG. Module 22, fourth determining module 23.
  • the second determining module 21 is configured to determine an original cross-delivery area order ratio corresponding to the plurality of historical orders in combination with an original delivery area corresponding to each of the plurality of historical orders.
  • the third determining module 22 is configured to determine, according to the delivery areas corresponding to the N clusters, a new cross-delivery area order ratio corresponding to the plurality of historical orders.
  • a fourth determining module 23 configured to set an order ratio according to the original cross-distribution area and the new cross-distribution area It is determined whether the distribution area corresponding to each of the N clusters is reasonable.
  • the apparatus shown in FIG. 8 can perform the method of the embodiment shown in FIG. 4.
  • the implementation process and technical effects of the technical solution refer to the description in the embodiment shown in FIG. 4, and details are not described herein again.
  • the structure of the distribution area dividing device can be implemented as an electronic device, such as a server, as shown in FIG. It may include a processor 31 and a memory 32.
  • the memory 32 is configured to store a program supporting the delivery area dividing means to execute the delivery area dividing method provided in any of the above embodiments, and the processor 31 is configured to execute the program stored in the memory 32.
  • the program includes one or more computer instructions, wherein the one or more computer instructions are for execution by the processor 31.
  • the processor 31 is configured to: perform clustering processing on the plurality of historical orders according to geographical similarity between the plurality of historical orders, to obtain N clusters, N ⁇ 1; according to the N clusters And a coverage area corresponding to each of the N clusters; and a corresponding distribution area of the N clusters according to the coverage areas corresponding to the N clusters.
  • the processor 31 is further configured to perform all or part of the foregoing method steps.
  • the structure of the order processing device may further include a communication interface 33 for the distribution area dividing device to communicate with other devices or communication networks.
  • an embodiment of the present disclosure provides a computer readable storage medium for storing computer software instructions for a distribution area dividing apparatus, including a program for performing a distribution area dividing method in each of the foregoing method embodiments.
  • the device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, ie may be located A place, or it can be distributed to multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the embodiment. Skill The general technicians in the field can understand and implement without any creative work.
  • the present disclosure discloses a method for dividing a delivery area, including:
  • the geographic location comprises a pickup location and a delivery location; the geographic similarity is measured by at least one of the following parameters: a distance between the pickup locations, and a delivery location The distance between the distance, the pick-up location, and the center point of the location of the shipping location.
  • A3 The method according to A2, the obtaining, according to a geographic location of a historical order included in each of the N clusters, a coverage area corresponding to each of the N clusters, including:
  • each of the N clusters corresponds to the historical order one by one, and each order line segment takes the picking position and the delivery position of the corresponding historical order as the end point;
  • each of the N clusters Obtaining a minimum closed polygon corresponding to each of the N clusters, wherein the minimum closed polygon causes an order line segment belonging to the corresponding cluster to be included in the minimum closed polygon, and each of the N clusters corresponds to a minimum closed polygon As the coverage area corresponding to each of the N clusters.
  • A4 The method according to A1, the determining, according to a coverage area corresponding to each of the N clusters, a distribution area corresponding to each of the N clusters, including:
  • A5. The method according to A1, the determining, according to a coverage area corresponding to each of the N clusters, a distribution area corresponding to each of the N clusters, including:
  • the cluster Ni and the cluster Nj are determined.
  • the number of order line segments corresponding in the overlapping area, j 1, 2,..., N, and j ⁇ i;
  • the distribution area corresponding to the cluster Ni is the coverage area corresponding to the adjusted cluster Ni.
  • the adjusting the coverage area corresponding to the cluster Ni according to the cluster to which the overlapping area belongs includes:
  • the method according to A6 the determining the M grids according to the number of order line segments respectively corresponding to the clusters Ni and the clusters Nj in each grid of the M grids After the respective class clusters, they also include:
  • the clusters to which the mesh Mk belongs are updated according to the clusters to which the plurality of adjacent meshes belong.
  • A8 The method of any of A1 to A7, further comprising:
  • the present disclosure discloses a distribution area dividing device, comprising:
  • a clustering module configured to cluster the plurality of historical orders according to geographical similarity between the plurality of historical orders, to obtain N clusters, N ⁇ 1;
  • An obtaining module configured to acquire, according to a geographic location of a historical order included in each of the N clusters, a coverage area corresponding to each of the N clusters;
  • a first determining module configured to determine, according to a coverage area corresponding to each of the N clusters, a delivery area corresponding to each of the N clusters.
  • the geographic location comprises a pickup location and a delivery location; the geographic similarity is measured by at least one of the following parameters: a distance between the pickup locations, and a delivery location The distance between the distance, the pick-up location, and the center point of the location of the shipping location.
  • a first acquiring unit configured to acquire, according to a geographic location of a historical order included in each of the N clusters, an order line segment corresponding to each of the N clusters, where the order line segment corresponds to a historical order, and each order line segment is The pick-up location and delivery location of the corresponding historical order are endpoints;
  • a second acquiring unit configured to acquire a minimum closed polygon corresponding to each of the N clusters, wherein the minimum closed polygon causes the order line segments belonging to the corresponding cluster to be included in the minimum closed polygon, the N classes
  • the smallest closed polygon corresponding to each of the clusters is used as a coverage area corresponding to each of the N clusters.
  • a first determining unit configured to: for any one of the N clusters, if the coverage area corresponding to the cluster Ni does not overlap with the coverage area corresponding to the other cluster Nj, determine the cluster
  • a third determining unit configured to determine, according to the number of the order line segments, a cluster of classes to which the overlapping area belongs;
  • an adjusting unit configured to adjust a coverage area corresponding to the cluster Ni according to the cluster to which the overlapping area belongs
  • the fourth determining unit is configured to determine that the delivery area corresponding to the cluster Ni is the coverage area corresponding to the adjusted cluster Ni.
  • the third determining unit is specifically configured to:
  • the adjusting unit is specifically configured to:
  • a recognition unit configured, for any one of the M grids, according to a position of the mesh Mk in the overlapping area and a class to which the plurality of adjacent meshes of the mesh Mk belong a cluster, identifying whether the mesh Mk is an abnormal mesh;
  • an updating unit configured to update the cluster to which the mesh Mk belongs according to the cluster to which the plurality of adjacent mesh belongs if the mesh Mk is an abnormal mesh.
  • a second determining module configured to determine an original cross-distribution area order ratio corresponding to the plurality of historical orders in combination with an original delivery area corresponding to each of the plurality of historical orders;
  • a third determining module configured to determine a new cross-delivery area order ratio corresponding to the plurality of historical orders in combination with a delivery area corresponding to each of the N clusters;
  • the fourth determining module is configured to determine, according to the original cross-distribution area order ratio and the new cross-delivery area order ratio, whether the delivery area corresponding to each of the N clusters is reasonable.
  • the disclosure also discloses C17, an electronic device, including a memory and a processor; wherein
  • the memory is for storing one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the delivery area partitioning method of any one of A1 to A8.
  • the present disclosure also discloses D18, a computer readable storage medium storing a computer program that, when executed by a computer, implements the distribution area dividing method according to any one of A1 to A8.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Human Resources & Organizations (AREA)
  • Marketing (AREA)
  • Development Economics (AREA)
  • General Business, Economics & Management (AREA)
  • General Engineering & Computer Science (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Databases & Information Systems (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Finance (AREA)
  • Remote Sensing (AREA)
  • Accounting & Taxation (AREA)
  • Game Theory and Decision Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

本公开实施例提供一种配送区域划分方法、装置、电子设备及计算机可读存储介质，该方法包括：根据多个历史订单彼此的地理位置相似度，对多个历史订单进行聚类处理，获得N个类簇；根据N个类簇各自包含的历史订单的地理位置，获取N个类簇各自对应的覆盖区域；根据N个类簇各自对应的覆盖区域确定N个类簇各自对应的配送区域。通过上述聚类处理可以使得不同类簇的覆盖区域相对独立，从而，基于对大量历史订单进行地理位置的聚类处理，实现对配送区域的自动划定，也可以有效降低跨配送区域订单的产生。

Description

配送区域划分方法、装置、电子设备及计算机可读存储介质 技术领域
本公开涉及互联网技术领域，尤其涉及一种配送区域划分方法、装置、电子设备及计算机可读存储介质。
背景技术
随着互联网的发展，线上到线下(Online To Offline，简称O2O)服务作为一种新型的服务模式，已经大大改变了人们的生活方式，比如购物方式，用户通过网上购物类应用，足不出户即可获取自己所需的物品。这些应用在便利用户的同时，也面临着订单配送问题，于是物流调度系统应运而生。
目前，在一些实际场景中，比如同城配送的场景中，物流调度系统对订单的调度往往是基于配送区域(一些场景中也称为商圈)进行调度的。简单来说，就是将某配送区域内产生的订单分配给归属于该配送区域的配送人员进行配送。
发明内容
以某城市来说，目前，对该城市进行配送区域的划分主要是相关工作人员凭借对该城市道路交通状况、商户分布情况、用户分布情况等的了解，在地图上经验划分出多个彼此覆盖区域不相交的配送区域。从而，在对该城市的订单进行调度时，则是基于人工划分得到的多个配送区域进行调度的。
人工经验划定的配送区域可能是不合理的，从而可能导致订单调度时，可能会导致配送运力的浪费。
举例来说，基于人工经验划定的配送区域可能产生大量的跨配送区域订单。在目前基于配送区域的订单调度机制下，这些跨配送区域订单会导致配送人员的空驶返程现象，是对配送运力的一种浪费。
以外卖订单的配送场景为例来说，订单中可以包含有取货位置和送货位置两种位置信息，那么当某订单的取货位置和送货位置对应于不同的配送区域时，该订单即为跨配送区域订单。
上述基于配送区域的订单调度机制，简单来说为：配送人员只能接受其所归属的配送区域的订单。其中，订单对应的配送区域可以这样确定：根据订单的取货位置所对应的配送区域确定该订单为哪个配送区域的订单。那么在对该订单进行调度时，需要将该订单分配给归属于配送区域A的某配送人员来完成订单的配送。而配送人员归属于哪个配送区域可以这样确定：根据配送人员在物流调度系统中提供的注册信息确定，也就是说，配送人员在注册时，即注册到了某个配送区域。基于此，假设某配送人员归属于配送区域A，其只能接受配送区域A的订单，意味着其只能接受取货位置在配送区域A的订单。
上述空驶返程现象可以根据如下举例理解：假设某个跨配送区域订单的取货位置对应于配送区域A，送货位置对应于配送区域B，此时，该订单被确定为是配送区域A的订单，会被分配给配送区域A中的某个配送人员来配送。该配送人员在行至配送区域B中的送货位置完成配送后，需要空驶返回到归属的配送区域A中继续接受配送区域A的订单，而不能在配送区域B内接受配送区域B内的订单，因为该配送人员归属于配送区域A，只能接受其归属的配送区域A的订单。从而，该跨配送区域订单造成了配送人员在返回至配送区域A的过程中的空跑现象，浪费了配送运力。
综上，由于人工经验有限、考虑因素单一，很可能导致配送区域范围的划分常有不合理的现象发生，对配送运力等会产生不利影响。
有鉴于此，本公开实施例提供一种配送区域划分方法、装置、电子设备及计算机可读存储介质，用以提高配送区域范围划分结果的合理性。
第一方面，本公开实施例提供一种配送区域划分方法，包括：
根据多个历史订单彼此的地理位置相似度，对所述多个历史订单进行聚类处理，获得N个类簇，N≥1；
根据所述N个类簇各自包含的历史订单的地理位置，获取所述N个类簇各自对应的覆盖区域；
根据所述N个类簇各自对应的覆盖区域确定所述N个类簇各自对应的配送区域。
第二方面，本公开实施例提供一种配送区域划分装置，包括：
聚类模块，用于根据多个历史订单彼此的地理位置相似度，对所述多个历史订单进行聚类处理，获得N个类簇，N≥1；
获取模块，用于根据所述N个类簇各自包含的历史订单的地理位置，获取所述N个类簇各自对应的覆盖区域；
第一确定模块，用于根据所述N个类簇各自对应的覆盖区域确定所述N个类簇各自对应的配送区域。
在一个可能的设计中，上述配送区域划分装置的结构中包括处理器和存储器，所述存储器用于存储支持配送区域划分装置执行上述第一方面中配送区域划分方法的程序，所述处理器被配置为用于执行所述存储器中存储的程序。所述配送区域划分装置还可以包括通信接口，用于配送区域划分装置与其他设备或通信网络通信。
第三方面，本公开实施例提供了一种计算机可读存储介质，用于储存配送区域划分装置所用的计算机软件指令，其包含用于执行上述第一方面中配送区域划分方法所涉及的程序。
本公开实施例提供的配送区域划分方法、装置、电子设备及计算机可读存储介质，获取大量实际产生的真实历史订单，通过结合这些历史订单彼此之间的地理位置相似度，对这些历史订单进行聚类处理，将地理位置更为相近的历史订单聚类到一个类簇中，从而获得N个类簇。进而，对于任一类簇来说，根据该类簇包含的历史订单的地理位置的分布，可以划定包含该类簇内全部历史订单的地理位置的一个闭合的覆盖区域，以据此覆盖区域确定该类簇所对应的配送区域的范围。由于属于同一类簇的历史订单彼此之间的地理位置相似度很高，而不同类簇间历史订单的地理位置相似度很低，因此，通过上述聚类处理可以使得不同类簇 的覆盖区域相对独立，从而，基于对大量历史订单进行地理位置的聚类处理，实现对配送区域的自动划定。
另外，本方案也可以被理解为是对各配送区域的范围进行不断优化的方案。假设执行本方案之前已经具有的各配送区域称为原配送区域，针对上述历史订单来说，基于各历史订单的地理位置在原配送区域的定位情况，假设其中有些历史订单属于跨配送区域订单。在经本方案的配送区域划分方法处理后，这些原来为跨配送区域的历史订单会被基于与其他历史订单的地理位置相似度聚类到某一类簇中，而该类簇的覆盖区域会包含该跨配送区域历史订单的地理位置，从而据此类簇的覆盖区域划定的配送区域的范围则会覆盖该跨配送区域历史订单的地理位置，那么后续再出现与该跨配送区域历史订单的地理位置相对应的新订单时，该新订单则不会再成为跨配送区域订单。从而说明，基于本方案对配送区域的不断优化划分，可以有效降低跨配送区域订单的产生。
附图说明
为了更清楚地说明本公开实施例或相关技术中的技术方案，下面将对实施例或相关技术描述中所需要使用的附图作一简单地介绍，显而易见地，下面描述中的附图是本公开的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为本公开实施例提供的配送区域划分方法实施例一的流程图；
图2为图1所示实施例中步骤102的一种可选实现方式的流程图；
图3a为图1所示实施例中步骤103的一种可选实现方式的流程图；
图3b-图3f为执行图3a所示实施例时的过程示意图；
图4为本公开实施例提供的配送区域划分方法实施例二的流程图；
图5为本公开实施例提供的配送区域划分装置实施例一的结构示意图；
图6为本公开实施例提供的配送区域划分装置实施例二的结构示意图；
图7为本公开实施例提供的配送区域划分装置实施例三的结构示意图；
图8为本公开实施例提供的配送区域划分装置实施例四的结构示意图；
图9为本公开实施例提供的与配送区域划分装置对应的电子设备的结构示意 图。
具体实施方式
为使本公开实施例的目的、技术方案和优点更加清楚，下面将结合本公开实施例中的附图，对本公开实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本公开一部分实施例，而不是全部的实施例。基于本公开中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本公开保护的范围。
在本公开实施例中使用的术语是仅仅出于描述特定实施例的目的，而非旨在限制本公开。在本公开实施例和所附权利要求书中所使用的单数形式的“一种”、“所述”和“该”也旨在包括多数形式，除非上下文清楚地表示其他含义，“多种”一般包含至少两种，但是不排除包含至少一种的情况。
应当理解，本文中使用的术语“和/或”仅仅是一种描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况。另外，本文中字符“/”，一般表示前后关联对象是一种“或”的关系。
应当理解，尽管在本公开实施例中可能采用术语第一、第二、第三等来描述XXX，但这些XXX不应限于这些术语。这些术语仅用来将XXX区分开。例如，在不脱离本公开实施例范围的情况下，第一XXX也可以被称为第二XXX，类似地，第二XXX也可以被称为第一XXX。
取决于语境，如在此所使用的词语“如果”、“若”可以被解释成为“在……时”或“当……时”或“响应于确定”或“响应于检测”。类似地，取决于语境，短语“如果确定”或“如果检测(陈述的条件或事件)”可以被解释成为“当确定时”或“响应于确定”或“当检测(陈述的条件或事件)时”或“响应于检测(陈述的条件或事件)”。
还需要说明的是，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的商品或者系统不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种商品或者系统所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的商品或者系统中还存在另外的相同要素。
进一步值得说明的是，本公开各实施例中各步骤之间的顺序是可以调整的，不是必须按照以下举例的顺序执行。
图1为本公开实施例提供的配送区域划分方法实施例一的流程图，本实施例提供的该配送区域划分方法可以由一配送区域划分装置来执行，该配送区域划分装置可以实现为软件，或者实现为软件和硬件的组合，该配送区域划分装置可以集成设置在物流调度平台侧的设备中，比如服务器中。如图1所示，该方法包括如下步骤：
101、根据多个历史订单彼此的地理位置相似度，对多个历史订单进行聚类处理，获得N个类簇，N≥1。
上述多个历史订单可以通过收集一定历史时间内、某城市的一个或多个原始配送区域所对应的订单获得。其中，该原始配送区域是指在执行本公开实施例提供的配送区域划分方法之前已经划得的配送区域。从这个角度来说，本公开实施例提供的配送区域划分方法也可以视为是一种配送区域划分范围的优化方法。
在一可选场景中，当实际需求是针对某城市中特定的一个或多个原始配送区域进行配送区域的重新划分处理时，基于订单所属配送区域的判定原则，即比如根据订单的取货位置所属的配送区域，在一定历史时间收到的全部历史订单中筛选出对应于上述一个或多个原始配送区域的全部或部分历史订单作为上述多个历史订单。
在订单配送领域中，订单所对应的地理位置可以包括取货位置和送货位置。以外卖场景来说，该取货位置往往对应于某商户的地址，该送货位置往往对应于 购买商品的用户的地址。
本实施例中，可选地，对多个历史订单进行聚类处理所依据的历史订单彼此的地理位置相似度，可以通过如下至少一种参数进行度量：取货位置之间的距离、送货位置之间的距离、取货位置与送货位置的位置中心点之间的距离。
可选地，对多个历史订单进行聚类处理，可以通过如下方式实现：预先设置相似度阈值，并计算任意两个历史订单间的地理位置相似度，从而将彼此的地理位置相似度大于或等于该相似度阈值的历史订单聚类到同一类簇中。具体实现时，在一种可选方式中，可以首先从多个历史订单中随机选择一个历史订单作为基准订单，该基准订单引出一个类簇，进而，基于剩余的未聚类历史订单分别与该基准订单间的地理位置相似度，从剩余的未聚类历史订单中选择出与该基准订单间的地理位置相似度大于相似度阈值的历史订单，加入到该基准订单对应的类簇中，从而形成与该基准订单对应的类簇。这里，未聚类历史订单是指还没有被聚类到一类簇内的历史订单。之后，针对剩余的未聚类历史订单，再进行上述基准订单的选择以及形成该基准订单对应的类簇的处理，直到所有的历史订单都被聚类到某个类簇中。
举例来说，假设上述多个历史订单包括订单1、订单2、订单3、订单4、订单5、订单6。第一次选择出的基准订单为订单1，引出类簇1，根据其他未聚类历史订单即订单2、订单3、订单4、订单5、订单6分别与订单1的地理位置相似度与相似度阈值的比较，假设订单2、订单3与订单1的地理位置相似度大于该相似度阈值，则将订单2、订单3聚类到订单1对应的类簇1中，从而，类簇1中包括订单1、订单2、订单3。之后，假设第二次选择的基准订单为订单6，引出类簇2，根据其他剩余订单即订单4、订单5分别与订单6的地理位置相似度与相似度阈值的比较，假设订单4与订单6的地理位置相似度大于该相似度阈值，则将订单4聚类到订单6对应的类簇2中，从而，类簇2中包括订单4、订单6。之后，只剩下订单5，订单5形成类簇3。
在另一种可选方式中，可以首先从多个历史订单中随机选择一个历史订单作为基准订单，引出一个类簇。进而，可以根据多个历史订单中的未聚类订单与该基准订单对应的类簇中的历史订单的地理位置相似度，对未聚类订单进行聚类处理。具体来说，该聚类处理过程是：每当向该类簇中加入一个历史订单后，下一个加入的历史订单需要满足这样的条件：与该类簇中已经加入的各历史订单间的地理位置相似度都大于或等于相似度阈值。
举例来说，假设上述多个历史订单包括订单1、订单2、订单3、订单4、订单5、订单6。第一次选择出的基准订单为订单1，引出类簇1。此时的未聚类历史订单为订单2、订单3、订单4、订单5、订单6，假设从中随机选择出订单2进行能否进入类簇1的判定，假设该订单2与订单1的地理位置相似度大于相似度阈值，被加入到类簇1中。此时，剩余的未聚类历史订单更新为订单3、订单4、订单5、订单6，进而，从该未聚类历史订单中选择出一个与类簇1中已加入的订单1和订单2之间的地理位置相似度均大于该相似度阈值的订单，假设订单5满足该条件，从而将订单5加入到类簇1中。假设此时剩余的未聚类历史订单即订单3、订单4、订单6中不存在与已加入类簇1中的各历史订单间的地理位置相似度均大于相似度阈值的历史订单，则最终形成的类簇1中包括订单1、订单2和订单5。之后，针对剩余的历史订单再进行上述的聚类处理，直到所有历史订单都被聚类到某个类簇中。
为了获得更佳的聚类效果，即为了使得属于同一类簇的历史订单彼此之间的地理位置相似度很高，而不同类簇间历史订单的地理位置相似度很低，实际应用中，上述地理位置相似度一般会通过多种度量参数的加权结果来表达。其中，在一种最简单的方式中，加权系数可以设置为1，也可以经验设定每种度量参数的加权系数。比如，地理位置相似度表达为：a/A+b/B，其中，A为取货位置之间的距离，B为送货位置之间的距离，加权系数a+b＝1。此时，意味着两个历史订单间的地理位置相似度需要考虑取货地址间距离以及送货地址间距离。可选地，可 以针对每种度量参数设置相应的相似度阈值，比如设置与取货地址间距离对应的阈值1，以及与送货地址间距离对应的阈值2，此时，两个历史订单的地理位置相似度由取货地址间距离对应的相似度分值1和送货地址间距离对应的相似度分值2组成，只有相似度分值1大于阈值1且相似度分值2大于阈值2时，该两个历史订单才被认为是较为相似的。其中，以相似度分值1为例，相似度分值1与其对应的取货地址间的距离之间具有预设的对应关系，比如可以通过预设函数式设定该对应关系，也就是说，可以预先设定取货地址间的距离与相似度分值之间的函数映射关系：f(L)，其中，L表示取货地址间的距离，f表示函数映射关系。
本实施例中，通过结合上述多个历史订单彼此之间的地理位置相似度，对这些历史订单进行聚类处理，假设可以获得N个类簇，其中，基于该聚类处理，地理位置更为相近的历史订单会被聚类到一个类簇中，即彼此间地理位置相似度大于或等于预设的相似度阈值的历史订单会被聚类到一个类簇中。
另外，假设基于原始配送区域来说，上述多个历史订单中的某历史订单为跨配送区域订单，即假设该历史订单的取货位置和收货位置分别对应于不同的原始配送区域。在经上述聚类处理后，该跨配送区域的历史订单可能会被基于与其他历史订单的地理位置相似度聚类到某一类簇中，因为多个历史订单中可能还会存在一定数量的历史订单与该历史订单一样，都是跨配送区域订单，且所跨的配送区域也是相同的。这也解释了为何说本公开实施例提供的配送区域划分方法也可以视为是一种配送区域划分范围的优化方法，因为如果两个原始配送区域间存在较多的跨配送区域订单，可能这两个原始配送区域的划分是不合适的。而如果某个类簇中包含了跨原始配送区域的历史订单，那么该类簇的覆盖区域会包含该跨原始配送区域的历史订单的取货地址和送货地址，从而在进行该类簇进行配送区域的划分时，新划分得到的配送区域将会覆盖该跨原始配送区域的历史订单的取货地址和送货地址，使得跨原始配送区域的历史订单在新划分得到的配送区域内不会再为跨配送区域的订单，从而，新划分得到的配送区域很可能会大大降低跨 配送区域订单的产生。
102、根据N个类簇各自包含的历史订单的地理位置，获取N个类簇各自对应的覆盖区域。
在对多个历史订单进行聚类处理，得到N个类簇后，这多个历史订单会被划分到不同的类簇中。其中，一个历史订单仅会被聚类到一个类簇中。进而，针对每个类簇来说，可选地，可以根据该类簇中包含的历史订单所对应的地理位置，在预设地图上划定该类簇对应的覆盖区域，该覆盖区域使得该类簇中全部历史订单所对应的地理位置都在该覆盖区域内。其中，该预设地图可以是用于进行上述某城市的配送区域划分之用的基础地图，其中包含有道路、预先设定的一些配送区域的名称，商户名称等元素，可选地，该预设地图上还可以显示有与各配送区域名称对应的原配送区域范围的标识。
在得到N个类簇，确定任一类簇对应的覆盖区域时，可以先在预设地图上，定位该任一类簇内包含的历史订单的标识，进而，在该预设地图上，划定该任一类簇的覆盖区域。
可选地，该历史订单的标识可以是某种形状的图案，比如圆点、方块、小红旗，用于标识对应历史订单的取货地址和送货地址，因此，每个历史订单可以具有两个标识，一个用于标识取货地址、一个用于标识送货地址。
可选地，任一类簇对应的覆盖区域的确定，可以通过采用凸包算法获得，即通过求取覆盖该类簇中全部历史订单所对应的地理位置的最小凸多边形来获得该覆盖区域。具体地，可以预先在预设地图上建立直角坐标系，针对任一类簇来说，将该任一类簇内包含的历史订单的标识定位到预设地图上之后，可以得到各历史订单标识在该坐标系中对应的坐标位置。进而，可以采用目前常用的凸包算法如Graham扫描法和Jarvis步进法进行各类簇对应的覆盖区域的求解。
103、根据N个类簇各自对应的覆盖区域确定N个类簇各自对应的配送区域。
本实施例中，在收集的上述多个历史订单足够充分的情况下，比如收集的历 史订单的数量至少达到一定数量阈值，和/或收集的天数已经至少达到一定的天数阈值，此时，可选地，针对N个类簇中任一类簇对应的配送区域来说，可以直接确定该任一类簇对应的配送区域为其对应的覆盖区域。或者，可选地，也可以确定该任一类簇对应的配送区域为其对应的覆盖区域与该类簇对应的原配送区域的并集。
针对确定该任一类簇对应的配送区域为其对应的覆盖区域与该类簇对应的原配送区域的并集来说，首先需要确定该类簇对应的原配送区域是哪个。具体地，因为基于前述基于地理位置相似度的聚类处理，聚入到一类簇中的历史订单基本上是由对应于同一原配送区域的历史订单组成，因此，某类簇对应的原配送区域可以根据该类簇中绝大多数历史订单所对应的原配送区域确定。
下面具体说明一下确定上述任一类簇对应的配送区域为其对应的覆盖区域与该类簇对应的原配送区域的并集后，相比于该原始配送区域，为何会降低跨配送区域订单的产生：
针对基于原始配送区域而言的某跨配送区域的历史订单来说，由于该跨配送区域的历史订单会被基于与其他历史订单的地理位置相似度聚类到某一类簇中，而该类簇的覆盖区域会包含该跨配送区域历史订单的地理位置，即包含其取货位置和送货位置，从而，根据该类簇的覆盖区域划定的配送区域的范围也会覆盖该跨配送区域历史订单的地理位置，那么后续再出现与该跨配送区域历史订单的地理位置相对应的新订单时，该新订单则不会再成为跨配送区域订单。从而说明，基于本实施例提供的配送区域划分方法对配送区域的不断优化划分，可以有效降低跨配送区域订单的产生。
综上，通过结合大量历史订单彼此之间的地理位置相似度，对该大量历史订单进行聚类处理，并基于各类簇所对应的覆盖区域范围确定对应类簇的配送区域，实现了配送区域的自动划定，并且可以使得划得的配送区域可以有利于降低跨配送区域订单的产生，从而有利于对配送运力的高效利用。
图2为图1所示实施例中步骤102的一种可选实现方式的流程图，如图2所示，可以包括如下步骤：
201、根据N个类簇各自包含的历史订单的地理位置，获取N个类簇各自对应的订单线段，订单线段与历史订单一一对应，每条订单线段以对应的历史订单的取货位置和送货位置为端点。
202、获取N个类簇各自对应的最小闭合多边形，该最小闭合多边形使得属于对应类簇的订单线段都被包含在该最小闭合多边形内，N个类簇各自对应的最小闭合多边形作为N个类簇各自对应的覆盖区域。
在得到N个类簇，确定任一类簇对应的覆盖区域时，可选地，可以在预设地图上先绘制该任一类簇内包含的历史订单的标识。进而，针对任一类簇来说，参见前述实施例中的说明，可以通过采用凸包算法等算法求取覆盖该类簇对应的全部订单线段的最小闭合多边形来获得该覆盖区域。
可选地，该历史订单的标识可以表现为订单线段的形式，此时，一个历史订单对应有一个订单线段，该订单线段的两个端点分别为该历史订单的取货位置和送货位置。
为了明显区分各类簇对应的订单线段，为方便进行后续的类簇对应的覆盖区域的确定，可以采用不同的绘制样式在预设地图上绘制各类簇对应的订单线段。其中，绘制样式的差别可以体现为颜色、线条形状等的不同。
上述借助预设地图来确定N个类簇各自对应的订单线段主要是可以在预设地图上真实地打点各历史订单对应的取货地址和送货地址，从而可以方便地获得各类簇对应的历史订单的订单线段分布情况。但是，也可以不借助于该预设地图实现。可选地，可以随机从多个历史订单中选择一个历史订单，以该历史订单的取货地址或送货地址作为参考地址，确定各历史订单的取货地址、送货地址相对该参考地址的相对位置，进而连接每个历史订单的取货地址和送货地址，得到每个历史订单的订单线段。
实际应用中，在基于一定算法获得各类簇对应的覆盖区域之后，在一可能的情况下，获得的覆盖区域并不一定是完全相互独立的覆盖区域，即可能部分覆盖区域存在重叠。下面，结合图3所示实施例对此情况下如何确定各类簇对应的配送区域。
图3a为图1所示实施例中步骤103的一种可选实现方式的流程图，如图3a所示，可以包括如下步骤：
301、对于N个类簇中的任一类簇Ni，判断类簇Ni对应的覆盖区域与其他类簇Nj对应的覆盖区域是否存在重叠，j＝1,2,…,N,且j≠i，若不重叠，则执行步骤302，否则，执行步骤303-306。
具体地，针对任意两个类簇来说，如果这两个类簇的覆盖区域存在如下的情况，则认为这两个类簇的覆盖区域存在重叠：如果存在至少一个订单线段，该订单线段对应的取货位置在这两个类簇的一个类簇中，该订单线段对应的送货位置在这两个类簇的另一个类簇中。
302、确定类簇Ni对应的配送区域为类簇Ni对应的覆盖区域。
对于任一类簇Ni来说，如果N个类簇中不存在覆盖区域与该任一类簇Ni的覆盖区域存在重叠的类簇，则可选地，该任一类簇Ni对应的配送区域可以直接确定为该任一类簇Ni对应的覆盖区域。
303、确定类簇Ni和类簇Nj在重叠区域内分别对应的订单线段个数。
304、根据类簇Ni和类簇Nj在重叠区域内分别对应的订单线段个数确定重叠区域归属的类簇。
305、根据重叠区域归属的类簇调整类簇Ni对应的覆盖区域。
306、确定类簇Ni对应的配送区域为调整后的类簇Ni对应的覆盖区域。
对于任一类簇Ni来说，如果N个类簇中至少存在某个类簇Nj，该类簇Nj的覆盖区域与类簇Ni的覆盖区域存在重叠，则可以通过确定该重叠区域的归属来调整类簇Ni的覆盖区域，同时意味着，类簇Nj的覆盖区域也会适应性调整。
该重叠区域的归属的确定，可以根据类簇Ni和类簇Nj分别落入该重叠区域内的订单线段的个数确定。其中，针对类簇Ni或类簇Nj来说，只要其对应的某订单线段与该重叠区域有交叉，即可以认为该订单线段落入该重叠区域，而并不一定要求该订单线段完全位于重叠区域内部才认为是落入该重叠区域。
对于落入该重叠区域内的某订单线段来说，由于该订单线段是用于表示一个历史订单的，该历史订单在经过聚类处理后被聚入了类簇Ni中或者类簇Nj中，从而，根据该订单线段对应的历史订单是被聚入了类簇Ni中还是类簇Nj中，可以确定该订单线段所对应的类簇。也就是说，在该重叠区域内，类簇Ni对应的订单线段即为落入该重叠区域内的，且对应的历史订单被聚入类簇Ni中的订单线段。
假设最终确定在该重叠区域内，类簇Ni对应的订单线段个数大于类簇Nj对应的订单线段个数，则可以认为该重叠区域归属于类簇Ni中，此时，类簇Ni的覆盖区域会扩大，而类簇Nj的覆盖区域可能由于需要去除重叠区域而会变小。相反地，假设最终确定在该重叠区域内，类簇Ni对应的订单线段个数小于类簇Nj对应的订单线段个数，则可以认为该重叠区域归属于类簇Nj中，此时，类簇Ni的覆盖区域由于需要去除重叠区域而会变小，而类簇Nj的覆盖区域可能扩大。
其中，以类簇Ni的覆盖区域会扩大，类簇Nj的覆盖区域会变小为例来说明此时类簇Ni和类簇Nj的覆盖区域的一种可选确定方式：由于将上述重叠区域归属给了类簇Ni，相当于将位于重叠区域内的订单线段所对应的历史订单都划至类簇Ni中了，此时，类簇Ni和类簇Nj分别包含的历史订单会发生更新：类簇Ni中会被加入重叠区域内的订单线段所对应的历史订单，相应的，类簇Nj中会被去除重叠区域内的订单线段所对应的历史订单，从而，可以根据类簇Ni和类簇Nj中更新后的历史订单的地理位置，调整类簇Ni和类簇Nj的覆盖区域。
结合图3b直观来看，图中，在类簇Ni与类簇Nj的覆盖区域的重叠区域中，假设订单线段a1、a2、a3为与类簇Ni对应的订单线段，订单线段b1和b2为与 类簇Nj对应的订单线段，则由于重叠区域中类簇Ni对应的订单线段个数大于类簇Nj对应的订单线段个数，从而，该重叠区域归属于类簇Ni，调整后的覆盖区域如图3c所示。
上述重叠区域的归属的确定，是将重叠区域作为一个整体来进行其归属的确定的。但是，实际上，可能会出现这样的情形：虽然类簇Ni和类簇Nj的覆盖区域存在重叠，但是，在重叠区域中，订单线段的分布可能呈现这样的特点：在更靠近类簇Ni的部分，类簇Ni对应的订单线段较多，在更靠近类簇Nj的部分，类簇Nj对应的订单线段较多。其中，对于位于类簇Ni和类簇Nj的重叠区域中的任一订单线段来说，更靠近类簇Ni往往意味着相对类簇Ni的覆盖区域边界的距离大于相对类簇Nj的覆盖区域边界的距离。此时，为了更加精确地确定重叠区域的归属，可以对重叠区域进行更小粒度的归属确定处理。
也就是说，对于上述重叠区域的归属的确定，既可以将该重叠区域作为一个整体进行其归属的确定，也可以将该重叠区域细分为多个子区域，通过确定多个子区域的归属来确定该重叠区域的归属。
以类簇Ni和类簇Nj的覆盖区域为例来说，可以先对该重叠区域进行网格划分，获得M个网格，如图3d所示。其中，网格的形状和大小可以根据实际需求而设定。值得说明的是，对该重叠区域进行网格划分既可以是直接针对该重叠区域进行网格划分实现的，也可以是通过对所有N个类簇的覆盖区域的整体进行网格划分而间接实现的。进而，针对M个网格的各个网格来说，确定类簇Ni和类簇Nj在各网格内分别对应的订单线段个数，以根据类簇Ni和类簇Nj在各网格内分别对应的订单线段个数，确定M个网格各自归属的类簇。其中，各网格所归属的类簇的确定过程可以参见前述说明理解，在此不赘述。如图3d所示，假设各网格的归属如图中所示，其中，网格中带有Ni字样，代表该网格归属于类簇Ni，网格中带有Nj字样，代表该网格归属于类簇Nj。
但是，值得说明的是，对于落入类簇Ni和类簇Nj的重叠区域中的任一订单 线段来说，该订单线段可能落入多个网格中，即与多个网格有交叉，比如对应于类簇Ni的某订单线段与网格1和网格2有交叉，那么，在进行类簇Ni在网格1内对应的订单线段个数的确定时，该订单线段被作为类簇Ni在网格1内对应的一条订单线段进行个数计算，而进行类簇Ni在网格2内对应的订单线段个数的确定时，该订单线段也会被作为类簇Ni在网格2内对应的一条订单线段进行个数计算。
在一可选场景中，如图3d所示，在确定M个网格各自归属的类簇之后，可能会出现这样的现象：对于其中的某个网格来说，该网格归属的类簇与其周围其他网格归属的类簇不同，此时，该网格的归属往往需要修正。
因此，对于M个网格中的任一网格Mk，可以根据网格Mk在重叠区域中的位置以及网格Mk的多个邻近网格所归属的类簇，识别网格Mk是否为异常网格；若网格Mk是异常网格，则可以根据网格Mk的多个邻近网格所归属的类簇，更新网格Mk所归属的类簇。如图3d中，差别显示出的网格，原来归属为类簇Nj，由于其周围网格大部分都归属于类簇Ni中，尤其是比其更加靠近类簇Nj的网格也归属于类簇Ni，则说明该网格为异常网格，需要进行归属类簇的修正，将其修正为类簇Ni，如图3e所示。导致这种现象出现的原因可能是：重叠区域内类簇Nj对应的订单线段恰好在某个网格的覆盖范围集中分布，而类簇Ni对应的订单线段在该网格的周围网格中广泛分布。
最后，可以根据M个网格各自归属的类簇调整类簇Ni对应的覆盖区域。具体地，将重叠区域内对应于类簇Ni的网格划入类簇Ni的覆盖区域中，将重叠区域内对应于类簇Nj的网格划入类簇Nj的覆盖区域中，如图3f所示。
从而，可选地，可以确定类簇Ni对应的配送区域为调整后的类簇Ni对应的覆盖区域。
以上实施例中，在对多个历史订单基于地理位置相似度进行聚类处理得到N个类簇后，当不同类簇之间的覆盖区域存在重叠时，通过对重叠区域进行准确的 归属类簇的确定，可以实现各类簇覆盖区域的合理调整，以使得基于该调整后的各类簇的覆盖区域确定出更为合理的配送区域。
图4为本公开实施例提供的配送区域划分方法实施例二的流程图，如图4所示，在前述各实施例基础上，以图1所示实施例为例，步骤103之后，还可以包括如下步骤：
401、结合多个历史订单各自对应的原始配送区域，确定多个历史订单对应的原跨配送区域订单比。
402、结合N个类簇各自对应的配送区域，确定多个历史订单对应的新跨配送区域订单比。
403、根据原跨配送区域订单比和新跨配送区域订单比确定N个类簇各自对应的配送区域是否合理。
在基于前述各实施例得到N个类簇各自对应的配送区域后，还可以验证N个类簇各自对应的配送区域划分的是否合理。可选地，由于不合理配送区域的划分可能会导致较多的跨配送区域订单的产生，因此，可以根据跨配送区域订单有无降低来验证N个类簇对应的配送区域的划分结果是否合理。
具体地，针对上述多个历史订单中的任一历史订单来说，可以根据其取货位置和送货位置是否位于同一个原始配送区域中来确定是否为跨配送区域订单，从而，在原始配送区域下，可以根据多个历史订单中跨配送区域订单的个数与该多个历史订单的总数的比值，确定多个历史订单对应的原跨配送区域订单比。另外，对于任一历史订单来说，由于经过聚类处理后该历史订单已经被聚入某一类簇中，根据该历史订单的取货位置和送货位置是否位于其聚入的类簇所对应的配送区域中，可以确定该历史订单当前是否为跨配送区域订单，从而，在新划分得到的N个类簇各自对应的配送区域下，可以根据多个历史订单中跨配送区域订单的个数与该多个历史订单的总数的比值，确定多个历史订单对应的新跨配送区域订单比。如果原跨配送区域订单比大于新跨配送区域订单比，说明新确定出的N个类 簇对应的配送区域合理，反之，说明新确定出的N个类簇对应的配送区域不合理。如果不合理，则可以保持各原始配送区域不变，即当前不对原始配送区域进行调整。
以下将详细描述本公开的一个或多个实施例的配送区域划分装置。本领域技术人员可以理解，这些配送区域划分装置均可使用市售的硬件组件通过本方案所教导的步骤进行配置来构成。
图5为本公开实施例提供的配送区域划分装置实施例一的结构示意图，如图5所示，该装置包括：聚类模块11、获取模块12、第一确定模块13。
聚类模块11，用于根据多个历史订单彼此的地理位置相似度，对所述多个历史订单进行聚类处理，获得N个类簇，N≥1。
获取模块12，用于根据所述N个类簇各自包含的历史订单的地理位置，获取所述N个类簇各自对应的覆盖区域。
第一确定模块13，用于根据所述N个类簇各自对应的覆盖区域确定所述N个类簇各自对应的配送区域。
其中，所述地理位置包括取货位置和送货位置；所述地理位置相似度通过如下至少一种参数度量：取货位置之间的距离、送货位置之间的距离、取货位置与送货位置的位置中心点之间的距离。
图5所示装置可以执行图1所示实施例的方法，本实施例未详细描述的部分，可参考对图1所示实施例的相关说明。该技术方案的执行过程和技术效果参见图1所示实施例中的描述，在此不再赘述。
图6为本公开实施例提供的配送区域划分装置实施例二的结构示意图，如图5所示，在图5所示实施例基础上，所述获取模块12包括：第一获取单元121、第二获取单元122。
第一获取单元121，用于根据所述N个类簇各自包含的历史订单的地理位置，获取所述N个类簇各自对应的订单线段，订单线段与历史订单一一对应，每条订 单线段以对应的历史订单的取货位置和送货位置为端点。
第二获取单元122，用于获取所述N个类簇各自对应的最小闭合多边形，所述最小闭合多边形使得属于对应类簇的订单线段都被包含在所述最小闭合多边形内，所述N个类簇各自对应的最小闭合多边形作为所述N个类簇各自对应的覆盖区域。
图6所示装置可以执行图2所示实施例的方法，本实施例未详细描述的部分，可参考对图2所示实施例的相关说明。该技术方案的执行过程和技术效果参见图2所示实施例中的描述，在此不再赘述。
图7为本公开实施例提供的配送区域划分装置实施例三的结构示意图，如图6所示，在前述实施例基础上，所述第一确定模块13包括：第一确定单元131、第二确定单元132、第三确定单元133、调整单元134、第四确定单元135。
在一种情形下：
第一确定单元131，用于对于所述N个类簇中的任一类簇Ni，若所述类簇Ni对应的覆盖区域与其他类簇Nj对应的覆盖区域不重叠，则确定所述类簇Ni对应的配送区域为所述类簇Ni对应的覆盖区域，j＝1,2,…,N,且j≠i。
在另一种情形下：
第二确定单元132，用于对于所述N个类簇中的任一类簇Ni，若所述类簇Ni对应的覆盖区域与其他类簇Nj对应的覆盖区域存在重叠，则确定所述类簇Ni和所述类簇Nj在重叠区域内分别对应的订单线段个数，j＝1,2,…,N,且j≠i。
第三确定单元133，用于根据所述订单线段个数确定所述重叠区域归属的类簇。
调整单元134，用于根据所述重叠区域归属的类簇调整所述类簇Ni对应的覆盖区域。
第四确定单元135，用于确定所述类簇Ni对应的配送区域为调整后的所述类簇Ni对应的覆盖区域。
可选地，所述第二确定单元132具体用于：对所述重叠区域进行网格划分，获得M个网格；确定所述类簇Ni和所述类簇Nj在所述M个网格的各网格内分别对应的订单线段个数。
相应地，所述第三确定单元133具体用于：根据所述类簇Ni和所述类簇Nj在所述M个网格的各网格内分别对应的订单线段个数，确定所述M个网格各自归属的类簇。
相应地，所述调整单元134具体用于：根据所述M个网格各自归属的类簇调整所述类簇Ni对应的覆盖区域。
可选地，所述第一确定模块13还包括：识别单元136、更新单元137。
识别单元136，用于对于所述M个网格中的任一网格Mk，根据所述网格Mk在所述重叠区域中的位置以及所述网格Mk的多个邻近网格所归属的类簇，识别所述网格Mk是否为异常网格。
更新单元137，用于若所述网格Mk是异常网格，则根据所述多个邻近网格所归属的类簇，更新所述网格Mk所归属的类簇。
图7所示装置可以执行图3a所示实施例的方法，本实施例未详细描述的部分，可参考对图3a所示实施例的相关说明。该技术方案的执行过程和技术效果参见图3a所示实施例中的描述，在此不再赘述。
图8为本公开实施例提供的配送区域划分装置实施例四的结构示意图，如图8所示，在图5所示实施例基础上，该装置还包括：第二确定模块21、第三确定模块22、第四确定模块23。
第二确定模块21，用于结合所述多个历史订单各自对应的原始配送区域，确定所述多个历史订单对应的原跨配送区域订单比。
第三确定模块22，用于结合所述N个类簇各自对应的配送区域，确定所述多个历史订单对应的新跨配送区域订单比。
第四确定模块23，用于根据所述原跨配送区域订单比和所述新跨配送区域订 单比确定所述N个类簇各自对应的配送区域是否合理。
图8所示装置可以执行图4所示实施例的方法，本实施例未详细描述的部分，可参考对图4所示实施例的相关说明。该技术方案的执行过程和技术效果参见图4所示实施例中的描述，在此不再赘述。
以上描述了配送区域划分装置的内部功能和结构，在一个可能的设计中，该配送区域划分装置的结构可实现为一电子设备，该电子设备比如为服务器，如图9所示，该电子设备可以包括：处理器31和存储器32。其中，所述存储器32用于存储支持配送区域划分装置执行上述任一实施例中提供的配送区域划分方法的程序，所述处理器31被配置为用于执行所述存储器32中存储的程序。
所述程序包括一条或多条计算机指令，其中，所述一条或多条计算机指令供所述处理器31调用执行。
所述处理器31用于：根据多个历史订单彼此的地理位置相似度，对所述多个历史订单进行聚类处理，获得N个类簇，N≥1；根据所述N个类簇各自包含的历史订单的地理位置，获取所述N个类簇各自对应的覆盖区域；根据所述N个类簇各自对应的覆盖区域确定所述N个类簇各自对应的配送区域。
可选地，所述处理器31还用于执行前述各方法步骤中的全部或部分步骤。
其中，所述订单处理装置的结构中还可以包括通信接口33，用于配送区域划分装置与其他设备或通信网络通信。
另外，本公开实施例提供了一种计算机可读存储介质，用于储存配送区域划分装置所用的计算机软件指令，其包含用于执行上述各方法实施例中配送区域划分方法所涉及的程序。
以上所描述的装置实施例仅仅是示意性的，其中所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。本领 域普通技术人员在不付出创造性的劳动的情况下，即可以理解并实施。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到各实施方式可借助加必需的通用硬件平台的方式来实现，当然也可以通过硬件。基于这样的理解，上述技术方案本质上或者说对相关技术做出贡献的部分可以以产品的形式体现出来，该计算机产品可以存储在计算机可读存储介质中，如ROM/RAM、磁碟、光盘等，包括若干指令用以使得一台计算机装置(可以是个人计算机，服务器，或者网络装置等)执行各个实施例或者实施例的某些部分所述的方法。
最后应说明的是：以上实施例仅用以说明本公开的技术方案，而非对其限制；尽管参照前述实施例对本公开进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本公开各实施例技术方案的精神和范围。
本公开公开A1、一种配送区域划分方法，包括：
根据多个历史订单彼此的地理位置相似度，对所述多个历史订单进行聚类处理，获得N个类簇，N≥1；
根据所述N个类簇各自包含的历史订单的地理位置，获取所述N个类簇各自对应的覆盖区域；
根据所述N个类簇各自对应的覆盖区域确定所述N个类簇各自对应的配送区域。
A2、根据A1所述的方法，所述地理位置包括取货位置和送货位置；所述地理位置相似度通过如下至少一种参数度量：取货位置之间的距离、送货位置之间的距离、取货位置与送货位置的位置中心点之间的距离。
A3、根据A2所述的方法，所述根据所述N个类簇各自包含的历史订单的地理位置，获取所述N个类簇各自对应的覆盖区域，包括：
根据所述N个类簇各自包含的历史订单的地理位置，获取所述N个类簇各自 对应的订单线段，订单线段与历史订单一一对应，每条订单线段以对应的历史订单的取货位置和送货位置为端点；
获取所述N个类簇各自对应的最小闭合多边形，所述最小闭合多边形使得属于对应类簇的订单线段都被包含在所述最小闭合多边形内，所述N个类簇各自对应的最小闭合多边形作为所述N个类簇各自对应的覆盖区域。
A4、根据A1所述的方法，所述根据所述N个类簇各自对应的覆盖区域确定所述N个类簇各自对应的配送区域，包括：
对于所述N个类簇中的任一类簇Ni，若所述类簇Ni对应的覆盖区域与其他类簇Nj对应的覆盖区域不重叠，则确定所述类簇Ni对应的配送区域为所述类簇Ni对应的覆盖区域，j＝1,2,…,N,且j≠i。
A5、根据A1所述的方法，所述根据所述N个类簇各自对应的覆盖区域确定所述N个类簇各自对应的配送区域，包括：
对于所述N个类簇中的任一类簇Ni，若所述类簇Ni对应的覆盖区域与其他类簇Nj对应的覆盖区域存在重叠，则确定所述类簇Ni和所述类簇Nj在重叠区域内分别对应的订单线段个数，j＝1,2,…,N,且j≠i；
根据所述订单线段个数确定所述重叠区域归属的类簇；
根据所述重叠区域归属的类簇调整所述类簇Ni对应的覆盖区域；
确定所述类簇Ni对应的配送区域为调整后的所述类簇Ni对应的覆盖区域。
A6、根据A5所述的方法，所述确定所述类簇Ni和所述类簇Nj在重叠区域内分别对应的订单线段个数，包括：
对所述重叠区域进行网格划分，获得M个网格；
确定所述类簇Ni和所述类簇Nj在所述M个网格的各网格内分别对应的订单线段个数；
所述根据所述订单线段个数确定所述重叠区域归属的类簇，包括：
根据所述类簇Ni和所述类簇Nj在所述M个网格的各网格内分别对应的订单 线段个数，确定所述M个网格各自归属的类簇；
所述根据所述重叠区域归属的类簇调整所述类簇Ni对应的覆盖区域，包括：
根据所述M个网格各自归属的类簇调整所述类簇Ni对应的覆盖区域。
A7、根据A6所述的方法，所述根据所述类簇Ni和所述类簇Nj在所述M个网格的各网格内分别对应的订单线段个数，确定所述M个网格各自归属的类簇之后，还包括：
对于所述M个网格中的任一网格Mk，根据所述网格Mk在所述重叠区域中的位置以及所述网格Mk的多个邻近网格所归属的类簇，识别所述网格Mk是否为异常网格；
若所述网格Mk是异常网格，则根据所述多个邻近网格所归属的类簇，更新所述网格Mk所归属的类簇。
A8、根据A1至A7中任一项所述的方法，所述方法还包括：
结合所述多个历史订单各自对应的原始配送区域，确定所述多个历史订单对应的原跨配送区域订单比；
结合所述N个类簇各自对应的配送区域，确定所述多个历史订单对应的新跨配送区域订单比；
根据所述原跨配送区域订单比和所述新跨配送区域订单比确定所述N个类簇各自对应的配送区域是否合理。
本公开公开B9、一种配送区域划分装置，包括：
聚类模块，用于根据多个历史订单彼此的地理位置相似度，对所述多个历史订单进行聚类处理，获得N个类簇，N≥1；
获取模块，用于根据所述N个类簇各自包含的历史订单的地理位置，获取所述N个类簇各自对应的覆盖区域；
第一确定模块，用于根据所述N个类簇各自对应的覆盖区域确定所述N个类簇各自对应的配送区域。
B10、根据B9所述的装置，所述地理位置包括取货位置和送货位置；所述地理位置相似度通过如下至少一种参数度量：取货位置之间的距离、送货位置之间的距离、取货位置与送货位置的位置中心点之间的距离。
B11、根据B10所述的装置，所述获取模块包括：
第一获取单元，用于根据所述N个类簇各自包含的历史订单的地理位置，获取所述N个类簇各自对应的订单线段，订单线段与历史订单一一对应，每条订单线段以对应的历史订单的取货位置和送货位置为端点；
第二获取单元，用于获取所述N个类簇各自对应的最小闭合多边形，所述最小闭合多边形使得属于对应类簇的订单线段都被包含在所述最小闭合多边形内，所述N个类簇各自对应的最小闭合多边形作为所述N个类簇各自对应的覆盖区域。
B12、根据B9所述的装置，所述第一确定模块包括：
第一确定单元，用于对于所述N个类簇中的任一类簇Ni，若所述类簇Ni对应的覆盖区域与其他类簇Nj对应的覆盖区域不重叠，则确定所述类簇Ni对应的配送区域为所述类簇Ni对应的覆盖区域，j＝1,2,…,N,且j≠i。
B13、根据B9所述的装置，所述第一确定模块包括：
第二确定单元，用于对于所述N个类簇中的任一类簇Ni，若所述类簇Ni对应的覆盖区域与其他类簇Nj对应的覆盖区域存在重叠，则确定所述类簇Ni和所述类簇Nj在重叠区域内分别对应的订单线段个数，j＝1,2,…,N,且j≠i；
第三确定单元，用于根据所述订单线段个数确定所述重叠区域归属的类簇；
调整单元，用于根据所述重叠区域归属的类簇调整所述类簇Ni对应的覆盖区域；
第四确定单元，用于确定所述类簇Ni对应的配送区域为调整后的所述类簇Ni对应的覆盖区域。
B14、根据B13所述的装置，所述第二确定单元具体用于：
对所述重叠区域进行网格划分，获得M个网格；确定所述类簇Ni和所述类 簇Nj在所述M个网格的各网格内分别对应的订单线段个数；
所述第三确定单元具体用于：
根据所述类簇Ni和所述类簇Nj在所述M个网格的各网格内分别对应的订单线段个数，确定所述M个网格各自归属的类簇；
所述调整单元具体用于：
根据所述M个网格各自归属的类簇调整所述类簇Ni对应的覆盖区域。
B15、根据B14所述的装置，所述第一确定模块还包括：
识别单元，用于对于所述M个网格中的任一网格Mk，根据所述网格Mk在所述重叠区域中的位置以及所述网格Mk的多个邻近网格所归属的类簇，识别所述网格Mk是否为异常网格；
更新单元，用于若所述网格Mk是异常网格，则根据所述多个邻近网格所归属的类簇，更新所述网格Mk所归属的类簇。
B16、根据B9至B15中任一项所述的装置，还包括：
第二确定模块，用于结合所述多个历史订单各自对应的原始配送区域，确定所述多个历史订单对应的原跨配送区域订单比；
第三确定模块，用于结合所述N个类簇各自对应的配送区域，确定所述多个历史订单对应的新跨配送区域订单比；
第四确定模块，用于根据所述原跨配送区域订单比和所述新跨配送区域订单比确定所述N个类簇各自对应的配送区域是否合理。
本公开还公开了C17、一种电子设备，包括存储器和处理器；其中，
所述存储器用于存储一条或多条计算机指令，其中，所述一条或多条计算机指令被所述处理器调用执行以实现如A1至A8中任一项所述的配送区域划分方法。
本公开还公开了D18、一种存储有计算机程序的计算机可读存储介质，所述计算机程序使计算机执行时实现如A1至A8中任一项所述的配送区域划分方法。

Claims (18)

  1. 一种配送区域划分方法，包括：
    根据多个历史订单彼此的地理位置相似度，对所述多个历史订单进行聚类处理，获得N个类簇，N≥1；
    根据所述N个类簇各自包含的历史订单的地理位置，获取所述N个类簇各自对应的覆盖区域；
    根据所述N个类簇各自对应的覆盖区域确定所述N个类簇各自对应的配送区域。
  2. 根据权利要求1所述的方法，其中，所述地理位置包括取货位置和送货位置；所述地理位置相似度通过如下至少一种参数度量：取货位置之间的距离、送货位置之间的距离、取货位置与送货位置的位置中心点之间的距离。
  3. 根据权利要求2所述的方法，其中，所述根据所述N个类簇各自包含的历史订单的地理位置，获取所述N个类簇各自对应的覆盖区域，包括：
    根据所述N个类簇各自包含的历史订单的地理位置，获取所述N个类簇各自对应的订单线段，其中，订单线段与历史订单一一对应，每条订单线段以对应的历史订单的取货位置和送货位置为端点；
    获取所述N个类簇各自对应的最小闭合多边形，所述最小闭合多边形使得属于对应类簇的订单线段都被包含在所述最小闭合多边形内，所述N个类簇各自对应的最小闭合多边形作为所述N个类簇各自对应的覆盖区域。
  4. 根据权利要求1所述的方法，其中，所述根据所述N个类簇各自对应的覆盖区域确定所述N个类簇各自对应的配送区域，包括：
    对于所述N个类簇中的任一类簇Ni，若所述类簇Ni对应的覆盖区域与其他类簇Nj对应的覆盖区域不重叠，则确定所述类簇Ni对应的配送区域为所述类簇Ni对应的覆盖区域，j＝1,2,…,N,且j≠i。
  5. 根据权利要求1所述的方法，其中，所述根据所述N个类簇各自对应的 覆盖区域确定所述N个类簇各自对应的配送区域，包括：
    对于所述N个类簇中的任一类簇Ni，若所述类簇Ni对应的覆盖区域与其他类簇Nj对应的覆盖区域存在重叠，则确定所述类簇Ni和所述类簇Nj在重叠区域内分别对应的订单线段个数，j＝1,2,…,N,且j≠i；
    根据所述订单线段个数确定所述重叠区域归属的类簇；
    根据所述重叠区域归属的类簇调整所述类簇Ni对应的覆盖区域；
    确定所述类簇Ni对应的配送区域为调整后的所述类簇Ni对应的覆盖区域。
  6. 根据权利要求5所述的方法，其中，所述确定所述类簇Ni和所述类簇Nj在重叠区域内分别对应的订单线段个数，包括：
    对所述重叠区域进行网格划分，获得M个网格；
    确定所述类簇Ni和所述类簇Nj在所述M个网格的各网格内分别对应的订单线段个数；
    所述根据所述订单线段个数确定所述重叠区域归属的类簇，包括：
    根据所述类簇Ni和所述类簇Nj在所述M个网格的各网格内分别对应的订单线段个数，确定所述M个网格各自归属的类簇；
    所述根据所述重叠区域归属的类簇调整所述类簇Ni对应的覆盖区域，包括：
    根据所述M个网格各自归属的类簇调整所述类簇Ni对应的覆盖区域。
  7. 根据权利要求6所述的方法，其中，所述根据所述类簇Ni和所述类簇Nj在所述M个网格的各网格内分别对应的订单线段个数，确定所述M个网格各自归属的类簇之后，还包括：
    对于所述M个网格中的任一网格Mk，根据所述网格Mk在所述重叠区域中的位置以及所述网格Mk的多个邻近网格所归属的类簇，识别所述网格Mk是否为异常网格；
    若所述网格Mk是异常网格，则根据所述多个邻近网格所归属的类簇，更新所述网格Mk所归属的类簇。
  8. 根据权利要求1至7中任一项所述的方法，其中，所述方法还包括：
    结合所述多个历史订单各自对应的原始配送区域，确定所述多个历史订单对应的原跨配送区域订单比；
    结合所述N个类簇各自对应的配送区域，确定所述多个历史订单对应的新跨配送区域订单比；
    根据所述原跨配送区域订单比和所述新跨配送区域订单比确定所述N个类簇各自对应的配送区域是否合理。
  9. 一种配送区域划分装置，包括：
    聚类模块，用于根据多个历史订单彼此的地理位置相似度，对所述多个历史订单进行聚类处理，获得N个类簇，N≥1；
    获取模块，用于根据所述N个类簇各自包含的历史订单的地理位置，获取所述N个类簇各自对应的覆盖区域；
    第一确定模块，用于根据所述N个类簇各自对应的覆盖区域确定所述N个类簇各自对应的配送区域。
  10. 根据权利要求9所述的装置，其中，所述地理位置包括取货位置和送货位置；所述地理位置相似度通过如下至少一种参数度量：取货位置之间的距离、送货位置之间的距离、取货位置与送货位置的位置中心点之间的距离。
  11. 根据权利要求10所述的装置，其中，所述获取模块包括：
    第一获取单元，用于根据所述N个类簇各自包含的历史订单的地理位置，获取所述N个类簇各自对应的订单线段，订单线段与历史订单一一对应，每条订单线段以对应的历史订单的取货位置和送货位置为端点；
    第二获取单元，用于获取所述N个类簇各自对应的最小闭合多边形，所述最小闭合多边形使得属于对应类簇的订单线段都被包含在所述最小闭合多边形内，所述N个类簇各自对应的最小闭合多边形作为所述N个类簇各自对应的覆盖区域。
  12. 根据权利要求9所述的装置，其中，所述第一确定模块包括：
    第一确定单元，用于对于所述N个类簇中的任一类簇Ni，若所述类簇Ni对应的覆盖区域与其他类簇Nj对应的覆盖区域不重叠，则确定所述类簇Ni对应的配送区域为所述类簇Ni对应的覆盖区域，j＝1,2,…,N,且j≠i。
  13. 根据权利要求9所述的装置，其中，所述第一确定模块包括：
    第二确定单元，用于对于所述N个类簇中的任一类簇Ni，若所述类簇Ni对应的覆盖区域与其他类簇Nj对应的覆盖区域存在重叠，则确定所述类簇Ni和所述类簇Nj在重叠区域内分别对应的订单线段个数，j＝1,2,…,N,且j≠i；
    第三确定单元，用于根据所述订单线段个数确定所述重叠区域归属的类簇；
    调整单元，用于根据所述重叠区域归属的类簇调整所述类簇Ni对应的覆盖区域；
    第四确定单元，用于确定所述类簇Ni对应的配送区域为调整后的所述类簇Ni对应的覆盖区域。
  14. 根据权利要求13所述的装置，其中，所述第二确定单元具体用于：
    对所述重叠区域进行网格划分，获得M个网格；确定所述类簇Ni和所述类簇Nj在所述M个网格的各网格内分别对应的订单线段个数；
    所述第三确定单元具体用于：
    根据所述类簇Ni和所述类簇Nj在所述M个网格的各网格内分别对应的订单线段个数，确定所述M个网格各自归属的类簇；
    所述调整单元具体用于：
    根据所述M个网格各自归属的类簇调整所述类簇Ni对应的覆盖区域。
  15. 根据权利要求14所述的装置，其中，所述第一确定模块还包括：
    识别单元，用于对于所述M个网格中的任一网格Mk，根据所述网格Mk在所述重叠区域中的位置以及所述网格Mk的多个邻近网格所归属的类簇，识别所述网格Mk是否为异常网格；
    更新单元，用于若所述网格Mk是异常网格，则根据所述多个邻近网格所归 属的类簇，更新所述网格Mk所归属的类簇。
  16. 根据权利要求9至15中任一项所述的装置，其中，还包括：
    第二确定模块，用于结合所述多个历史订单各自对应的原始配送区域，确定所述多个历史订单对应的原跨配送区域订单比；
    第三确定模块，用于结合所述N个类簇各自对应的配送区域，确定所述多个历史订单对应的新跨配送区域订单比；
    第四确定模块，用于根据所述原跨配送区域订单比和所述新跨配送区域订单比确定所述N个类簇各自对应的配送区域是否合理。
  17. 一种电子设备，包括存储器和处理器；其中，
    所述存储器用于存储一条或多条计算机指令，其中，所述一条或多条计算机指令被所述处理器调用执行以实现如权利要求1至8中任一项所述的配送区域划分方法。
  18. 一种存储有计算机程序的计算机可读存储介质，其中，所述计算机程序使计算机执行时实现如权利要求1至8中任一项所述的配送区域划分方法。
PCT/CN2017/109999 2017-06-14 2017-11-08 配送区域划分方法、装置、电子设备及计算机可读存储介质 WO2018227859A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US16/713,318 US20200118075A1 (en) 2017-06-14 2019-12-13 Method and apparatus for dividing delivery regions, electronic device, and computer-readable storage medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710449213.6 2017-06-14
CN201710449213.6A CN107451673B (zh) 2017-06-14 2017-06-14 配送区域划分方法和装置

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/713,318 Continuation US20200118075A1 (en) 2017-06-14 2019-12-13 Method and apparatus for dividing delivery regions, electronic device, and computer-readable storage medium

Publications (1)

Publication Number Publication Date
WO2018227859A1 true WO2018227859A1 (zh) 2018-12-20

Family

ID=60486531

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/109999 WO2018227859A1 (zh) 2017-06-14 2017-11-08 配送区域划分方法、装置、电子设备及计算机可读存储介质

Country Status (3)

Country Link
US (1) US20200118075A1 (zh)
CN (1) CN107451673B (zh)
WO (1) WO2018227859A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111126688A (zh) * 2019-12-19 2020-05-08 北京顺丰同城科技有限公司 配送路线确定方法、装置、电子设备及可读存储介质
CN112561412A (zh) * 2019-09-10 2021-03-26 顺丰科技有限公司 目标对象标识的确定方法、装置、服务器及存储介质
CN114770538A (zh) * 2022-04-24 2022-07-22 国网上海市电力公司 一种机器人自动巡检方法

Families Citing this family (37)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108009653B (zh) * 2017-08-16 2021-12-21 北京嘀嘀无限科技发展有限公司 订单管理方法、装置、服务器和计算机可读存储介质
CN109919167B (zh) * 2017-12-12 2021-07-06 北京京东乾石科技有限公司 分拣中心的货物分拣方法和装置、货物分拣系统
CN108133406B (zh) * 2017-12-21 2021-10-15 北京星选科技有限公司 信息显示方法、装置、电子设备及计算机可读存储介质
CN110110950A (zh) * 2018-02-01 2019-08-09 北京京东振世信息技术有限公司 生成配送路区的方法、装置及计算机可读存储介质
CN110210797B (zh) * 2018-02-28 2023-11-07 北京京东尚科信息技术有限公司 生产覆盖区域确定方法和装置以及计算机可读存储介质
CN110363453B (zh) * 2018-03-26 2022-11-08 北京京东振世信息技术有限公司 配送信息可视化方法和装置
CN110414758B (zh) * 2018-04-28 2021-11-30 北京三快在线科技有限公司 区域确定方法、装置及电子设备
CN110503352B (zh) * 2018-05-16 2021-09-07 北京三快在线科技有限公司 一种确定交付点的方法、装置及计算机可读存储介质
CN110503353B (zh) * 2018-05-16 2022-04-01 北京三快在线科技有限公司 一种配送区域表达方法及装置
CN110555448B (zh) * 2018-05-30 2022-03-29 顺丰科技有限公司 一种细分派件区域的方法和系统
CN110659849B (zh) * 2018-06-28 2021-07-06 北京三快在线科技有限公司 业务覆盖区域的确定方法、装置及电子设备
CN108985694B (zh) * 2018-07-17 2022-04-22 北京百度网讯科技有限公司 用于确定配送中心地址的方法和装置
CN109064218B (zh) * 2018-07-17 2021-04-27 北京三快在线科技有限公司 划分区域的方法、装置及电子设备
CN109003028B (zh) * 2018-07-17 2022-05-27 北京百度网讯科技有限公司 用于划分物流区域的方法和装置
CN109242387A (zh) * 2018-09-07 2019-01-18 上海墨盾电脑科技有限公司 一种供应链实现方法
CN109447319B (zh) * 2018-09-26 2023-06-16 中国平安财产保险股份有限公司 一种网格划分方法、计算机可读存储介质及终端设备
CN109598430B (zh) * 2018-11-28 2021-01-08 拉扎斯网络科技（上海）有限公司 配送范围生成方法、装置、电子设备和存储介质
CN109615137A (zh) * 2018-12-13 2019-04-12 合肥工业大学智能制造技术研究院 用于云物流环境下云配送的选址优化方法
CN109657027B (zh) * 2018-12-19 2020-11-03 金瓜子科技发展（北京）有限公司 一种聚类选址的方法、装置、存储介质及电子设备
CN109670721A (zh) * 2018-12-26 2019-04-23 拉扎斯网络科技（上海）有限公司 任务调度方法、装置、电子设备及计算机可读存储介质
CN111754147A (zh) * 2019-03-28 2020-10-09 北京京东尚科信息技术有限公司 路区划分方法、系统、装置和计算机可读存储介质
CN110071832B (zh) * 2019-04-18 2022-04-15 中国联合网络通信集团有限公司 通信质量保障方法和装置
CN110111054A (zh) * 2019-05-09 2019-08-09 上汽安吉物流股份有限公司 拼单网络模型的生成方法及装置、计算机可读介质以及物流系统
CN111915115A (zh) * 2019-05-10 2020-11-10 北京沃东天骏信息技术有限公司 执行策略设置方法和装置
CN112053105A (zh) * 2019-06-05 2020-12-08 北京京东振世信息技术有限公司 划分服务区域的方法和装置
CN110414613B (zh) * 2019-07-31 2021-03-02 京东城市(北京)数字科技有限公司 区域聚类的方法、装置、设备和计算机可读存储介质
CN110969387A (zh) * 2019-11-28 2020-04-07 拉扎斯网络科技（上海）有限公司 订单配送方法、服务器、终端及系统
CN112948512A (zh) * 2019-12-10 2021-06-11 顺丰科技有限公司 位置数据划分方法、装置、计算机设备和存储介质
CN111242165B (zh) * 2019-12-30 2020-10-23 北京顺达同行科技有限公司 商户聚类方法、装置、计算机设备和存储介质
CN111340364A (zh) * 2020-02-26 2020-06-26 上海东普信息科技有限公司 同城订单配送范围绘制方法及装置
CN111461779B (zh) * 2020-03-31 2023-10-24 拉扎斯网络科技（上海）有限公司 地图信息处理的方法、装置、可读存储介质和电子设备
CN111461627B (zh) * 2020-04-27 2023-09-26 上海东普信息科技有限公司 快递员派送区域的划分方法、装置、设备及存储介质
CN111768161B (zh) * 2020-07-14 2023-11-21 上海燕汐软件信息科技有限公司 一种跨营业部派送统计方法、装置及系统
CN111915256B (zh) * 2020-07-31 2023-09-26 上海寻梦信息技术有限公司 构建派件围栏的方法、异地签收识别方法及相关设备
CN112215580B (zh) * 2020-10-23 2024-02-06 岭东核电有限公司 核电作业区域设置方法、装置、计算机设备和存储介质
US20230105216A1 (en) * 2021-10-05 2023-04-06 Argo Al, LLC System and Method for Generating a Planned Path for a Vehicle Using a Cloud Deployment System
CN114372754B (zh) * 2022-01-11 2023-04-28 拉扎斯网络科技（上海）有限公司 订单匹配方法、装置及计算机设备

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014163977A1 (en) * 2013-03-13 2014-10-09 Google Inc. Systems, methods and computer-readable media for interpreting geographical search queries
CN104102953A (zh) * 2014-06-24 2014-10-15 四川省烟草公司广安市公司 一种物流送货线路优化生成方法及系统
CN105825360A (zh) * 2016-03-31 2016-08-03 北京小度信息科技有限公司 商户配送范围的调整方法和装置
CN105868843A (zh) * 2016-03-22 2016-08-17 南京邮电大学 一种面向货品配送的路线规划方法

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104077308B (zh) * 2013-03-28 2018-02-13 阿里巴巴集团控股有限公司 一种物流服务范围确定方法及装置
CN104123305B (zh) * 2013-04-28 2018-08-24 国际商业机器公司 地理数据处理方法及其系统
CN104200369B (zh) * 2014-08-27 2019-12-31 北京京东尚科信息技术有限公司 一种确定商品配送范围的方法和装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014163977A1 (en) * 2013-03-13 2014-10-09 Google Inc. Systems, methods and computer-readable media for interpreting geographical search queries
CN104102953A (zh) * 2014-06-24 2014-10-15 四川省烟草公司广安市公司 一种物流送货线路优化生成方法及系统
CN105868843A (zh) * 2016-03-22 2016-08-17 南京邮电大学 一种面向货品配送的路线规划方法
CN105825360A (zh) * 2016-03-31 2016-08-03 北京小度信息科技有限公司 商户配送范围的调整方法和装置

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112561412A (zh) * 2019-09-10 2021-03-26 顺丰科技有限公司 目标对象标识的确定方法、装置、服务器及存储介质
CN112561412B (zh) * 2019-09-10 2022-07-08 顺丰科技有限公司 目标对象标识的确定方法、装置、服务器及存储介质
CN111126688A (zh) * 2019-12-19 2020-05-08 北京顺丰同城科技有限公司 配送路线确定方法、装置、电子设备及可读存储介质
CN111126688B (zh) * 2019-12-19 2023-05-26 北京顺丰同城科技有限公司 配送路线确定方法、装置、电子设备及可读存储介质
CN114770538A (zh) * 2022-04-24 2022-07-22 国网上海市电力公司 一种机器人自动巡检方法
CN114770538B (zh) * 2022-04-24 2023-12-19 国网上海市电力公司 一种机器人自动巡检方法

Also Published As

Publication number Publication date
US20200118075A1 (en) 2020-04-16
CN107451673A (zh) 2017-12-08
CN107451673B (zh) 2019-05-14

Similar Documents

Publication Publication Date Title
WO2018227859A1 (zh) 配送区域划分方法、装置、电子设备及计算机可读存储介质
TWI598839B (zh) Method and apparatus for determining a target location
US11895515B2 (en) Systems and methods for communications node upgrade and selection
TWI584137B (zh) Search, determine the active area of ​​the method with the server
CN108415952B (zh) 用户数据存储方法、标签计算方法及计算设备
JP6162781B2 (ja) 新しい興味のある地点の位置を決定する方法及び装置
CN107767092B (zh) 商品对象信息的处理方法、展示方法及装置
US20160036718A1 (en) Network service analytics
CN107330761B (zh) 基于地理位置的服务信息提供方法和装置以及计算设备
CN109741142A (zh) 订单分配方法、订单分配装置、可读存储介质和电子设备
CN111859187B (zh) 基于分布式图数据库的poi查询方法、装置、设备及介质
CN110020925A (zh) 订单处理方法、装置、设备及存储介质
CN110401552A (zh) 一种cdn业务自动调优方法及装置
CN107943423A (zh) 云系统中存储资源的管理方法和计算机可读存储介质
US20150120367A1 (en) Geospatially relevant supply chain optimization through a networked mobile device that leverages a distributed computing environment
US10489724B2 (en) Predicting resources required by a customer and their cost
CN108182589A (zh) 商圈辐射范围确定方法及装置
CN110428627B (zh) 一种公交出行潜力区域识别方法及识别系统
US20180144289A1 (en) Inventory demand forecasting system and inventory demand forecasting method
CN108537654B (zh) 客户关系网络图的渲染方法、装置、终端设备及介质
JP6749967B2 (ja) 情報処理装置、方法及びプログラム
CN113190587A (zh) 实现业务数据处理的数据处理方法和装置
US20210019710A1 (en) Computing system that is configured to infer locations of enterprise rooms
CN104363470B (zh) 一种流服务测试方法和系统
CN117892984B (zh) 运力调度方法、程序产品、设备及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17913445

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17913445

Country of ref document: EP

Kind code of ref document: A1