WO2020258756A1 - 配送方适合混合送模式的确定 - Google Patents

配送方适合混合送模式的确定 Download PDF

Info

Publication number
WO2020258756A1
WO2020258756A1 PCT/CN2019/125168 CN2019125168W WO2020258756A1 WO 2020258756 A1 WO2020258756 A1 WO 2020258756A1 CN 2019125168 W CN2019125168 W CN 2019125168W WO 2020258756 A1 WO2020258756 A1 WO 2020258756A1
Authority
WO
WIPO (PCT)
Prior art keywords
distribution range
distributor
delivery
range
distribution
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2019/125168
Other languages
English (en)
French (fr)
Inventor
郝小菠
张润丰
丁雪涛
何仁清
郭振刚
韩菁
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Sankuai Online Technology Co Ltd
Original Assignee
Beijing Sankuai Online Technology Co Ltd
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 Beijing Sankuai Online Technology Co Ltd filed Critical Beijing Sankuai Online Technology Co Ltd
Publication of WO2020258756A1 publication Critical patent/WO2020258756A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G06Q10/0832—Special goods or special handling procedures, e.g. handling of hazardous or fragile goods
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G06Q10/0835—Relationships between shipper or supplier and carriers
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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

Definitions

  • This application relates to the field of logistics technology, and in particular to the determination of a delivery party suitable for a mixed delivery mode.
  • special delivery can refer to the mode of delivery by full-time delivery personnel of the logistics platform.
  • Fast delivery may refer to a mode in which social personnel perform part-time delivery when they are free (fast delivery is also called crowdsourcing in some embodiments).
  • Special delivery has the advantages of good service experience, but also has the disadvantages of high delivery cost and small delivery range.
  • fast delivery has the advantages of low distribution cost and large distribution range, but also has the disadvantage of poor service experience.
  • the distribution platform will open a mixed delivery mode of special delivery and fast delivery for the delivery party (hereinafter referred to as the mixed delivery mode), that is, use special delivery within a preset smaller distribution range .
  • the mixed delivery mode In order to ensure the service experience; in a larger range, that is, a distribution range outside the smaller distribution range, use fast delivery to reduce distribution costs and increase distribution orders.
  • the present application provides a method, device, computer storage medium and electronic equipment for determining that a distributor is suitable for a mixed delivery mode.
  • a method for determining that a distributor is suitable for a mixed delivery mode includes:
  • the second distribution range of the target distributor is generated based on the first distribution range of the target distributor; wherein, the second distribution range surrounds the first distribution range, and the target distributor has not yet activated the mixed delivery mode And intend to open the mixed delivery mode;
  • a device for identifying that a distributor is suitable for a mixed delivery mode comprising:
  • a feature obtaining unit which determines a second distribution range feature based on the second distribution range, and obtains the first distribution range feature of the first distribution range and the target distribution party's distribution party feature;
  • a prediction unit based on the second distribution range feature, the first distribution range feature, and the distributor feature, predicting the change value of the business in the second distribution range after the target distributor adopts the mixed delivery mode;
  • the determining unit determines that the target delivery party is suitable for the mixed delivery mode when the change value of the business in the second delivery range meets the preset condition.
  • a computer-readable storage medium stores a computer program, and the computer program is used to execute any one of the above methods for identifying a delivery party suitable for a mixed delivery mode.
  • An electronic device including:
  • a memory for storing processor executable instructions
  • the processor is configured to execute any one of the foregoing methods for identifying that the distributor is suitable for a mixed delivery mode.
  • the embodiment of the present application provides a solution for identifying that a distributor is suitable for a mixed delivery mode, and generating a second distribution range of the target distributor according to the first distribution range of the target distributor to be identified; according to the second distribution range Determine the characteristics of the second distribution range, and obtain the characteristics of the first distribution range of the first distribution range and the characteristics of the dispatcher of the target distributor; according to the characteristics of the first distribution range, the characteristics of the second distribution range, and the characteristics of the dispatcher , Predicting the change value of the business in the second distribution range after the target dispatcher adopts the mixed delivery mode; when the change value of the business in the second distribution range meets the preset condition, determine the target dispatcher Suitable for mixed delivery mode.
  • the prediction result is based on the objectively existing characteristics of the first distribution range, the second distribution range and the distribution party, the prediction result is also objective and reliable.
  • Fig. 1 is a schematic diagram of a first distribution range and a second distribution range shown in an exemplary embodiment of the present application;
  • Fig. 2 is a flowchart of a method for determining that a distributor is suitable for a mixed delivery mode according to an exemplary embodiment of the present application
  • Fig. 3 is a hardware structure diagram of a device for determining that a distributor is suitable for a mixed delivery mode according to an exemplary embodiment of the present application
  • Fig. 4 is a block diagram of a device for determining that a distributor is suitable for a mixed delivery mode according to an exemplary embodiment of the present application.
  • first, second, third, etc. may be used in this application to describe various information, the information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other.
  • first information may also be referred to as second information, and similarly, the second information may also be referred to as first information.
  • word “if” as used herein can be interpreted as "when” or “when” or "in response to determination”.
  • the distribution range diagram shown in FIG. 1 shows the distribution range of the special delivery mode described in an embodiment of the present application (hereinafter referred to as the first distribution range) and the distribution range of the express mode outside the first distribution range (Hereinafter referred to as the second delivery range).
  • the location of the merchant is P (the position of the black dot in the figure).
  • P the position of the black dot in the figure.
  • special delivery is used within a preset smaller delivery range ; Use fast delivery in a larger outer ring range. Therefore, the first delivery range of the merchant in FIG. 1 is S1 (the range of the white circular area in the figure); and the S2 (the black ring area in the figure) surrounding the first delivery range S1 is the present embodiment
  • the mixed delivery model is not suitable for all delivery parties.
  • the opening of the mixed delivery mode may cause the distributor to actually lose money in the newly added second distribution range.
  • the distribution platform reviews whether the distributor is suitable for the mixed delivery mode mainly depends on the human experience of the auditors, and lacks a plan to objectively evaluate whether the distributor is suitable for the mixed delivery mode.
  • this application provides a solution to determine that the distributor is suitable for the mixed delivery mode.
  • By predicting the business change of the distributor in the second distribution range after adopting the mixed delivery mode it is judged whether the distributor is suitable for the mixed delivery.
  • the following takes an instant delivery scenario as an example to introduce the delivery party and other related business parties described in this application.
  • the instant delivery scenario may include, but is not limited to, offline on-site services such as takeaway, express delivery, and errands.
  • merchants and customers can use the instant delivery APP to fulfill logistics orders, the customer can be regarded as the logistics recipient, the instant delivery APP can be regarded as the server, and the merchant can be regarded as the dispatcher.
  • Customers can browse different merchants and a variety of products provided by merchants through the instant delivery app.
  • the customer After the customer determines the product and the merchant, they can place an order; generally, during the ordering process, the customer needs to provide logistics information such as receiving Address, contact information, etc.; after the customer places an order, the server can push the order information to the corresponding merchant. The merchant needs to confirm whether the order is accepted. If the order is confirmed, the server can perform the delivery based on the delivery mode of the merchant. member.
  • the server can match the optimal delivery courier for delivery; when the customer's receiving address is in the merchant's express delivery mode When it corresponds to the second distribution range, the server can match the optimal express delivery staff for distribution; when the customer’s receiving address is located in both the first distribution range and the second distribution range of the merchant (that is, the outermost of the first distribution range) And when it is located at the innermost part of the second distribution range), based on the cost-priority strategy (the delivery fee charged by express delivery personnel is lower than that of dedicated delivery personnel), the server can match the optimal express delivery personnel for delivery.
  • the following is a flow chart of a method for determining that a delivery party is suitable for a mixed delivery mode shown in conjunction with FIG. 2 in an exemplary embodiment of the present application.
  • the method may be applied to a server, and the method may include the following steps:
  • Step 110 Generate a second distribution range of the target distributor based on the first distribution range of the target distributor; wherein, the second distribution range surrounds the first distribution range, and the target distributor does not open a mixed delivery mode And intends to open the mixed delivery mode.
  • the server may receive a request from a target distributor, and the request carries the first distribution range data of the target distributor.
  • both the first delivery range and the second delivery range may refer to coordinate data represented by a number of position coordinates (for example, latitude and longitude coordinates).
  • the original delivery range boundary can be planned based on the location coordinates (for example, a convex hull algorithm can be used to draw the maximum range of the area contained in these location coordinates on the map).
  • the target distributor Since the current target distributor has not yet opened the mixed delivery mode, the target distributor has only the first distribution range, but not the second distribution range. Therefore, the server needs to generate a second distribution range for the target distributor based on the first distribution range.
  • the server may use a preset second distribution range generation algorithm to generate a second distribution range that surrounds the first distribution range outside the first distribution range.
  • the server may extend a predetermined distance outward from the boundary of the first delivery range, and determine the area extended by the predetermined distance as the second delivery range.
  • the boundary coordinate points of the first delivery range are used as reference points, and each reference point is respectively expanded outward by a preset distance to obtain the boundary coordinate points of the second delivery range;
  • a convex hull algorithm may be used to connect the boundary coordinate points of the plurality of second distribution ranges into a closed curve of a convex hull.
  • the range outside the first distribution range is the second distribution range.
  • the preset distance may be a preset empirical value.
  • the preset distance may be fixed; it may also be dynamically changed, for example, it may be flexibly adjusted according to the specific situation of the area where the business is located.
  • the preset distance can be adjusted according to the current time period in the area where the business is located.
  • different time periods have different delivery requirements.
  • the preset distance can be shortened accordingly in order to prevent merchants from breaking orders; conversely, when the delivery demand is low
  • the preset distance can be increased accordingly.
  • the preset distance can be adjusted according to the current capacity of the area where the merchant is located. Generally, when the capacity is tight, the preset distance can be shortened correspondingly in order to avoid exploding orders; conversely, when the capacity is sufficient, the preset distance can be correspondingly longer.
  • the server may determine an area within a preset distance outside the first distribution range as the second distribution range according to preset parameters.
  • the newly obtained range after the expansion is the second distribution range.
  • the newly obtained range after the expansion is the second distribution range.
  • this embodiment does not need to perform complicated calculations and can generate the second distribution range simply and quickly.
  • the first delivery range described in this application can be circular, polygonal or any other shape.
  • the expanded second distribution range can also be of any shape. As shown in Figure 1, the second distribution range is a circular distribution range.
  • the currently generated second distribution range is actually a virtual second distribution range, which is not actually applicable to the target distributor. That is to say, although the server generates the second distribution range of the target distributor, the second distribution range is invisible to the target distributor or any other user, and does not participate in or affect the target distributor. Current actual business.
  • Step 120 Determine a second distribution range characteristic based on the second distribution range, and obtain the first distribution range characteristic of the first distribution range and the target dispatcher's dispatcher characteristic.
  • the server After the server generates the second distribution range of the target distributor, it may determine the second distribution range feature of the target distributor based on the second distribution range.
  • the characteristics of the first distribution range may include one or more of the following characteristics: historical order distribution of the delivery party in the first distribution range, historical distribution experience and other characteristics, historical order revenue, and the first distribution range Other characteristics related to internal profit and loss;
  • the second distribution range characteristics may include one or more of the following characteristics: the user's demand characteristics in the second distribution range, including user preferences, user consumption amount, and other characteristics related to profit and loss in the second distribution range;
  • the characteristics of the dispatcher may include one or more of the following characteristics: basic information of the dispatcher, the surrounding capacity of the dispatcher, the characteristics of the comparison structure with the surrounding dispatchers, and the characteristics of the comparison structure between the first distribution range and the second distribution range.
  • Step 130 Based on the characteristics of the second distribution range, the characteristics of the first distribution range, and the characteristics of the distributor, predict the change value of the business in the second distribution range after the target distributor adopts the mixed delivery mode.
  • the server determines the second distribution range feature of the target distributor, and obtains the first distribution range feature of the first distribution range and the target distributor's distributor characteristics, it can predict that the target distributor uses hybrid The change value of the business within the second distribution range after the delivery mode.
  • the server may pre-train a machine learning model for predicting the business change value in the second distribution range after the target distributor adopts the hybrid delivery mode based on the machine learning technology.
  • the step 130 includes:
  • the server may pre-collect a large number of historical characteristics of the second distribution range, historical characteristics of the first distribution range, and historical characteristics of the distributor generated by a large number of distributors that have enabled the mixed delivery mode, as well as the historical characteristics of each distributor in the second distribution.
  • the historical change value of the business within the scope.
  • the historical characteristics of the second distribution range, the historical characteristics of the first distribution range, the historical characteristics of the distributor and the historical change value (paired data) of each distributor are used as the training sample set (generally, the characteristics can be used as the input value, and the change Value as the output value), and train the machine learning model based on the machine learning algorithm.
  • the machine learning model can be continuously improved.
  • the exit conditions for example, the accuracy of the predicted change value and the actual change value for all delivery parties meets the business requirements
  • the target distributor’ When predicting the change value of the target distributor’s business, the target distributor’s second distribution range characteristics, first distribution range characteristics, and distributor characteristics are input as input data to the machine learning model, and the machine learning model can be obtained according to training The algorithm calculates and outputs a business change value within the second distribution range.
  • the machine learning model may adopt, for example, logistic regression (LR), gradient boosting decision tree (GBDT), random forest algorithm, etc.
  • LR logistic regression
  • GBDT gradient boosting decision tree
  • random forest algorithm etc.
  • the algorithm adopted by the machine learning model may be:
  • A represents the change value of the business within the second distribution range
  • B represents the first distribution range feature
  • C represents the second distribution range feature
  • D represents the delivery party feature.
  • the machine learning model using the random forest algorithm has the advantages of better model generalization performance and fast model training convergence.
  • Step 140 In the case that the change value of the business in the second distribution range meets the preset condition, determine that the target distributor is suitable for the mixed delivery mode.
  • the preset conditions may be flexibly set in advance according to business requirements.
  • the preset conditions can also be set by the target delivery party.
  • different target distributors can have different preset conditions to adapt to the individual needs of different distributors and achieve the effect of thousands of people.
  • the change value of the business in the second distribution range may include the profit value of the target distributor in the second distribution area.
  • the preset condition may be related to the profit value.
  • the target distributor when the change value, that is, the profit value reaches (equal to or greater than) a threshold, it is determined that the target distributor is suitable for the mixed delivery mode.
  • the threshold value can also be preset by the target distributor.
  • the merchant in order to ensure profitability, can set the threshold to a positive value; in this way, the merchant is allowed to open the mixed delivery mode only when the business in the second distribution range is profitable.
  • the merchant in order to increase popularity or traffic, can also set the threshold to a negative value, that is, after the merchant opens the hybrid delivery mode, the business within the second distribution range can be allowed to lose money; as long as the loss amount does not exceed the threshold Both can open the mixed delivery mode.
  • the target delivery party can be flexibly set based on business requirements in actual applications.
  • the change value of the business in the second distribution range may include the delivery person's pick-up time, and the preset condition may be less than the preset time; that is, when the delivery person's pick-up time is less than the preset time In this case, it is determined that the target distributor is suitable for the mixed delivery mode. This can ensure that the delivery staff will not take too long to pick up the goods and avoid delivery overtime.
  • the change value of the business in the second distribution range may include the receiving time of the logistics receiver, and the preset condition may be less than the preset time; that is, the receiving time of the logistics receiving party is less than the preset time. In this case, it is determined that the target distributor is suitable for the mixed delivery mode. This can ensure that the receiving time of the logistics receiver will not be too long and avoid delivery timeout.
  • the above-mentioned variation values can be used in combination to more accurately determine whether the target distributor is suitable for the mixed delivery mode.
  • the server may enable the mixed delivery mode for the target distributor.
  • the server may enable the mixed delivery mode for the target distributor, and validate the second distribution range of the target distributor generated above, so that the target distributor’s
  • the actual business can cover the second distribution range. For example, push the target distributor to users in the second distribution range, so that users in the second distribution range can see the services provided by the target distributor.
  • this application provides a solution for determining that a delivery party is suitable for a mixed delivery mode, determining a second delivery range feature according to the second delivery range, and obtaining the first delivery range feature and the first delivery range of the first delivery range.
  • Distributor characteristics of the target distributor according to the second distribution range characteristics, the first distribution range characteristics, and the distributor characteristics, predict business changes in the second distribution range after the target distributor adopts the mixed delivery mode Value; in the case that the change value of the business in the second distribution range meets the preset conditions, it is determined that the target distributor is suitable for the mixed delivery mode.
  • the prediction result is determined based on the objectively existing characteristics of the first distribution range, the second distribution range and the characteristics of the distributor, the prediction result is also objectively consistent with the laws of nature.
  • this application also provides an embodiment of the device for determining that the distributor is suitable for the mixed delivery mode.
  • the embodiment of the device for determining that the distributor is suitable for the mixed delivery mode in this application can be applied to the server.
  • the device embodiments can be implemented by software, or by hardware or a combination of software and hardware. Taking software implementation as an example, as a logical device, it is formed by reading the corresponding computer program instructions in the non-volatile memory into the memory through the processor where it is located.
  • this application identifies a hardware structure diagram of a device suitable for the mixed delivery mode of the distributor, except for the processor, memory, network interface, and non-volatile memory shown in Figure 3
  • the embodiment usually determines the actual function of the distributor to be suitable for the mixed delivery mode according to the actual function of the delivery party, and may also include other hardware, which will not be repeated here.
  • the device for determining that the distributor is suitable for the mixed delivery mode may include:
  • the generating unit 310 generates a second distribution range of the target distributor based on the first distribution range of the target distributor; wherein, the second distribution range surrounds the first distribution range, and the target distributor has not opened mixed delivery Mode and intends to open the mixed delivery mode;
  • a feature acquisition unit 320 which determines a second distribution range feature based on the second distribution range, and obtains the first distribution range feature of the first distribution range and the target distributor's shipper feature;
  • a prediction unit 330 based on the second distribution range feature, the first distribution range feature, and the distributor feature, predicting the change value of the business in the second distribution range after the target distributor adopts the mixed delivery mode;
  • the determining unit 340 determines that the target delivery party is suitable for the mixed delivery mode when the change value of the business in the second delivery range meets the preset condition.
  • the generating unit 310 includes:
  • the boundary of the first distribution range is extended outward by a preset distance, and the area extended by the preset distance is determined as the second distribution range.
  • the generating unit 310 includes:
  • the expansion subunit takes the boundary coordinate points of the first delivery range as the reference point, and each of the reference points respectively expands outwards by a preset distance to obtain the boundary coordinate points of the second delivery range;
  • a sub-unit is generated to determine the boundary of the second distribution range according to the navigation path between adjacent boundary coordinate points of the boundary coordinate points of the second distribution range; wherein, the boundary of the second distribution range is the same as that of the first distribution.
  • the area between the boundaries of the range is the second delivery range.
  • the prediction unit 330 includes:
  • the output subunit obtains the change value of the business within the second distribution range after the target distributor adopts the mixed distribution mode and output after calculation by the machine learning model.
  • the device further includes the following subunits for training the preset machine learning model:
  • a training subunit which inputs the historical characteristics of the second distribution range, the historical characteristics of the first distribution range, the historical characteristics of the distributor, and the historical change value of the business within the second distribution range as training samples into the machine learning algorithm training; wherein, The historical characteristics of the second distribution range, the historical characteristics of the first distribution range, and the historical characteristics of the distributor are used as sample input, and the historical change value of the second distribution range business is output as a sample;
  • the machine learning algorithm includes a random forest algorithm.
  • the determining unit 340 includes:
  • the characteristics of the first distribution range include:
  • At least one of user preferences or user consumption amount within the second distribution range At least one of user preferences or user consumption amount within the second distribution range
  • At least one of the basic information of the distributor or the surrounding capacity of the distributor At least one of the basic information of the distributor or the surrounding capacity of the distributor.
  • the relevant part can refer to the part of the description of the method embodiment.
  • the device embodiments described above are merely illustrative.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in One place, or it can be distributed to multiple network units.
  • Some or all of the modules can be selected according to actual needs to achieve the objectives of the solution of this application. Those of ordinary skill in the art can understand and implement it without creative work.
  • the above Figure 4 depicts the internal functional modules and structural diagrams of the device for identifying the distributor suitable for the mixed delivery mode.
  • the substantial execution body may be an electronic device, including:
  • a memory for storing processor executable instructions
  • the processor is configured to:
  • the second distribution range of the target distributor is generated based on the first distribution range of the target distributor; wherein, the second distribution range surrounds the first distribution range, and the target distributor does not open a mixed delivery mode and intends Open mixed delivery mode;
  • the generating the second distribution range of the target distributor based on the first distribution range of the target distributor includes:
  • the boundary of the first distribution range is extended outward by a preset distance, and the area extended by the preset distance is determined as the second distribution range.
  • the extension of the boundary of the first distribution range outward by a preset distance, and determining the area extended by the preset distance as the second distribution range includes:
  • each reference point extends outwards by a preset distance to obtain the boundary coordinate points of the second delivery range;
  • predicting the change value of the business in the second distribution range after the target distributor adopts the mixed delivery mode includes:
  • it also includes: training the preset machine learning model in the following manner:
  • the historical characteristics of the second distribution range, the historical characteristics of the first distribution range, the historical characteristics of the distributor and the historical change value of the business in the second distribution range are input as training samples to the machine learning algorithm training; wherein, the second The historical characteristics of the distribution range, the historical characteristics of the first distribution range and the historical characteristics of the distributor are used as sample inputs, and the historical change value of the business in the second distribution range is output as a sample;
  • the machine learning algorithm includes a random forest algorithm.
  • the determining that the target distributor is suitable for a mixed delivery mode in the case that the change value of the business within the second distribution range meets a preset condition includes:
  • the target distributor is suitable for a mixed delivery mode.
  • the characteristics of the first distribution range include:
  • At least one of user preferences or user consumption amount within the second distribution range At least one of user preferences or user consumption amount within the second distribution range
  • At least one of the basic information of the distributor or the surrounding capacity of the distributor At least one of the basic information of the distributor or the surrounding capacity of the distributor.
  • the processor may be a central processing unit (English: Central Processing Unit, abbreviated as: CPU), or other general-purpose processors, digital signal processors (English: Digital Signal Processor) , Abbreviation: DSP), Application Specific Integrated Circuit (English: Application Specific Integrated Circuit, Abbreviation: ASIC), etc.
  • the general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.
  • the aforementioned memory can be read-only memory (English: read-only memory, abbreviation: ROM), random access memory (English : Random access memory, referred to as RAM), flash memory, hard disk or solid state disk.
  • the steps of the method disclosed in the embodiments of the present invention may be directly embodied as being executed and completed by a hardware processor, or executed and completed by a combination of hardware and software modules in the processor.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Economics (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Marketing (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Strategic Management (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Development Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Software Systems (AREA)
  • Mathematical Physics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

一种配送方适合混合送模式的确定方案系统。其包括：基于目标配送方的第一配送范围生成所述目标配送方的第二配送范围；其中，所述第二配送范围包含所述第一配送范围，所述目标配送方暂未开通所述混合送模式且有意向开通所述混合送模式（110）；基于所述第二配送范围确定第二配送范围特征，以及获取所述第一配送范围的第一配送范围特征和所述目标配送方的配送方特征（120）；基于所述第二配送范围特征、第一配送范围特征和配送方特征，预测所述目标配送方采用混合送模式后所述第二配送范围内业务的变化值（130）；在所述第二配送范围内业务的变化值符合预设条件的情况下，确定所述目标配送方适合混合送模式（140）。

Description

配送方适合混合送模式的确定 技术领域
本申请涉及物流技术领域，尤其涉及配送方适合混合送模式的确定。
背景技术
在物流配送中，通常存在专送和快送两种不同配送模式。其中，专送可以是指由物流平台的专职配送员进行配送的模式。快送可以是指由社会人员在空闲的时候兼职进行配送的模式(在有的实施例中快送也会称为众包)。专送具有服务体验好的优点，但也存在配送成本高、配送范围小的缺点。而，快送具有配送成本低、配送范围大的优点，也存在服务体验差的缺点。
通常，配送平台为了平衡服务体验和配送成本，会为配送方开通专送和快送混合的配送模式(以下简称为混合送模式)，即在预设的一个较小的配送范围内使用专送，以确保服务体验；在更大的范围内，即该较小的配送范围之外的一个配送范围使用快送，以降低配送成本、提升配送订单量。
发明内容
有鉴于此，本申请提供一种确定配送方适合混合送模式的方法、装置及计算机存储介质和电子设备。
具体地，本申请是通过如下技术方案实现的：
一种确定配送方适合混合送模式的方法，所述方法包括：
基于目标配送方的第一配送范围生成所述目标配送方的第二配送范围；其中，所述第二配送范围包围所述第一配送范围，所述目标配送方暂未开通所述混合送模式且有意向开通所述混合送模式；
基于所述第二配送范围确定第二配送范围特征，以及获取所述第一配送范围的第一配送范围特征和所述目标配送方的配送方特征；
基于所述第二配送范围特征、第一配送范围特征和配送方特征，预测所述目标配送方采用混合送模式后所述第二配送范围内业务的变化值；
在所述第二配送范围内业务的变化值符合预设条件的情况下，确定所述目标配送方适合混合送模式。
一种识别配送方适合混合送模式的装置，所述装置包括：
生成单元，基于目标配送方的第一配送范围生成所述目标配送方的第二配送范围；其中，所述第二配送范围包围所述第一配送范围，所述目标配送方未开通所述混合送模式且有意向开通所述混合送模式；
特征获取单元，基于所述第二配送范围确定第二配送范围特征，以及获取所述第一配送范围的第一配送范围特征和所述目标配送方的配送方特征；
预测单元，基于所述第二配送范围特征、第一配送范围特征和配送方特征，预测所述目标配送方采用混合送模式后所述第二配送范围内业务的变化值；
确定单元，在所述第二配送范围内业务的变化值符合预设条件的情况下，确定所述目标配送方适合混合送模式。
一种计算机可读存储介质，所述存储介质存储有计算机程序，所述计算机程序用于执行上述任一项识别配送方适合混合送模式的方法。
一种电子设备，包括：
处理器；
用于存储处理器可执行指令的存储器；
所述处理器被配置为执行上述任一项识别配送方适合混合送模式的方法。
本申请实施例，提供了一种识别配送方适合混合送模式的方案，根据待识别的目标配送方的第一配送范围生成所述目标配送方的第二配送范围；根据所述第二配送范围确定第二配送范围特征，以及获取所述第一配送范围的第一配送范围特征和所述目标配送方的配送方特征；根据所述第一配送范围特征、第二配送范围特征和配送方特征，预测所述目标配送方采用混合送模式后所述第二配送范围内业务的变化值；在所述第二配送范围内业务的变化值符合预设条件的情况下，确定所述目标配送方适合混合送模式。如此，通过预测配送方在采用混合送模式后第二配送范围内业务的变化情况判断该配送方是否适合进行混合送，从而可以仅为适合进行混合送的配送方开通混合送模式。由于预 测结果是基于客观存在的第一配送范围特征、第二配送范围特征以及配送方特征，因此预测结果也是客观可靠的。
附图说明
图1是本申请一示例性实施例示出的第一配送范围和第二配送范围的示意图；
图2是本申请一示例性实施例示出的一种确定配送方适合混合送模式的方法的流程图；
图3是本申请一示例性实施例示出的一种确定配送方适合混合送模式的装置的硬件结构图；
图4是本申请一示例性实施例示出的一种确定配送方适合混合送模式的装置的框图。
具体实施方式
这里将详细地对示例性实施例进行说明，其示例表示在附图中。下面的描述涉及附图时，除非另有表示，不同附图中的相同数字表示相同或相似的要素。以下示例性实施例中所描述的实施方式并不代表与本申请相一致的所有实施方式。相反，它们仅是与如所附权利要求书中所详述的、本申请的一些方面相一致的装置和方法的例子。
在本申请使用的术语是仅仅出于描述特定实施例的目的，而非旨在限制本申请。在本申请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。还应当理解，本文中使用的术语“和/或”是指并包含一个或多个相关联的列出项目的任何或所有可能组合。
应当理解，尽管在本申请可能采用术语第一、第二、第三等来描述各种信息，但这些信息不应限于这些术语。这些术语仅用来将同一类型的信息彼此区分开。例如，在不脱离本申请范围的情况下，第一信息也可以被称为第二信息，类似地，第二信息也可以被称为第一信息。取决于语境，如在此所使用的词语“如果”可以被解释成为“在……时”或“当……时”或“响应于确定”。
图1所示的配送范围示意图示出了本申请一实施例中所述的专送模式的配送范围 (以下简称为第一配送范围)，以及第一配送范围之外的快送模式的配送范围(以下简称为第二配送范围)。
图1中，商家的位置为P(图中黑色圆点的位置)，当该商家采用混合送模式时，基于前述混合送模式的特点，在预设的一个较小的配送范围内使用专送；在更大的外环范围内使用快送。因此，图1中该商家的第一配送范围为S1(图中白色圆形区域的范围)；而包围在所述第一配送范围S1之外的S2(图中黑色环形范围)即本实施例中所述的第二配送范围。
然而，混合送模式并非适合所有的配送方。在有的情况下，开通混合送模式后可能会导致配送方在新增的第二配送范围内实际是亏损的。通常，配送平台审核配送方是否适合开通混合送模式主要依靠审核人员的人为经验，而缺乏客观评价配送方是否适合混合送模式的方案。
为此本申请提供了一种确定配送方适合混合送模式的方案，通过预测配送方在采用混合送模式后第二配送范围内业务的变化情况判断该配送方是否适合进行混合送，从而可以仅为适合进行混合送的配送方开通混合送模式。由于预测结果是基于客观存在的第一配送范围特征、第二配送范围特征以及配送方特征而确定的，因此预测结果也是客观可靠的。
以下以即时配送场景为例，介绍本申请中所述的配送方以及其它相关的业务方。所述即时配送场景可以包括但不限于外卖、快递、跑腿等线下上门服务。在即时配送场景中，商家和顾客可以通过即时配送APP达成物流订单，顾客即可以为物流接收方，即时配送APP即可以为服务器，商家即可以为配送方。顾客通过即时配送APP可以浏览不同的商家以及商家提供的多种多样的商品，当顾客确定了商品以及所在的商家后，就可以进行下单；一般下单过程中，需要顾客提供物流信息如接收地址、联系方式等；在顾客下单后，服务器就可以推送下单信息给对应的商家，商家需要确认是否接单，如果确认接单则可以由服务器基于该商家的配送模式匹配进行配送的配送员。如果商家采用混合送模式，当顾客的接收地址位于商家的专送模式对应的第一配送范围时，服务器可以匹配最优的专送配送员进行配送；当顾客的接收地址位于商家的快送模式对应的第二配送范围时，服务器可以匹配最优的快送配送员进行配送；当顾客的接收地址同时位于商家的第一配送范围和第二配送范围内(即位于第一配送范围的最外侧且位于第二配送范围的最内侧)时，基于成本优先的策略(快送配送员收取的配送费低于专送配送员)，服务器可以匹配最优的快送配送员进行配送。
以下结合图2本申请一示例性实施例示出的一种确定配送方适合混合送模式的方法流程图，所述方法可以应用于服务器中，该方法可以包括如下步骤：
步骤110：基于目标配送方的第一配送范围生成所述目标配送方的第二配送范围；其中，所述第二配送范围包围所述第一配送范围，所述目标配送方未开通混合送模式且有意向开通混合送模式。
本申请实施例中，服务器可以接收目标配送方的请求，该请求中携带有所述目标配送方的第一配送范围数据。一般的，不管是第一配送范围还是第二配送范围都可以是指由若干的位置坐标(例如经纬度坐标)表示的坐标数据。通常，基于位置坐标可以规划出原始的配送范围的边界(例如采用凸包算法可以在地图上画出这些位置坐标所包含的最大范围的区域)。
由于当前目标配送方还未开通混合送模式，因此目标配送方只有第一配送范围，而没有第二配送范围。所以，服务器需要基于该第一配送范围为目标配送方生成第二配送范围。
服务器可以采用预设的第二配送范围生成算法，在该第一配送范围之外生成一个包围第一配送范围的第二配送范围。
在一实施例中，服务器可以以第一配送范围的边界向外扩展预设距离，将所述扩展预设距离的区域确定为第二配送范围。
具体地，以构建第一配送范围的边界坐标点为基准点，将每个基准点分别向外扩展预设距离得到第二配送范围的边界坐标点；
根据所述多个第二配送范围的边界坐标点中相邻的边界坐标点之间的导航路径，确定第二配送范围的边界；其中，所述第二配送范围的边界与第一配送范围的边界之间的区域为第二配送范围。
在一实施例中，确定第二配送范围的边界，可以采用凸包算法，将所述多个第二配送范围的边界坐标点连接成一个凸包的闭合曲线，所述凸包的闭合曲线内、所述第一配送范围外的范围即为第二配送范围。
在一实施例中，所述预设距离可以是预先设置的经验值。所述预设距离可以是固定的；也可以是动态变化的，例如可以根据商家所在区域的特定情况灵活进行调整。
例如，可以根据商家所在区域的当前时间段，对预设距离进行调整。一般的，不同 时间段对配送需求也不同，在配送需求旺盛的时间段如午餐时间段和晚餐时间段，为了不使商家爆单可以将预设距离相应缩短；反之，在配送需求较低的时间段如凌晨时间段，可以将预设距离相应变长。
再例如，可以根据商家所在区域的当前运力情况，对预设距离进行调整。一般的，在运力紧张时，为了避免爆单可以将预设距离相应缩短；反之，在运力充足时，可以将预设距离相应变长。
在一实施例中，服务器可以按照预设参数，将第一配送范围之外的预设距离内的区域确定为第二配送范围。
例如，在第一配送范围的基础上向外扩大1倍，那么向外扩大后新得到的范围即为第二配送范围。
再例如，在第一配送范围的基础上向外扩大3公里，那么向外扩大后新得到的范围即为第二配送范围。
相比于前一实施例，该实施例无需进行复杂计算，可以简单快速地生成第二配送范围。
本申请中所述的第一配送范围可以是圆形、多边形或者其它任意形状。而经过扩展后的第二配送范围同样可以是任意形状的。如图1中，第二配送范围为一个环形的配送范围。
值得一提的是，由于目标配送方当前还未开通混合送模式，因此当前生成的第二配送范围实际是虚拟的第二配送范围，并非目标配送方真实可以应用的。也就是说，虽然服务器生成了目标配送方的第二配送范围，但该第二配送范围对于目标配送方或者其它任何用户来说都是不可见不可感知的，也不参与、不影响目标配送方当前的实际业务。
步骤120：基于所述第二配送范围确定第二配送范围特征，以及获取所述第一配送范围的第一配送范围特征和所述目标配送方的配送方特征。
服务器在生成了目标配送方的第二配送范围后，可以基于该第二配送范围确定目标配送方的第二配送范围特征。
其中，所述第一配送范围特征可以包括以下特征中的一个或多个：配送方在第一配送范围内历史的订单分布、历史的配送体验等特征、历史订单的收益，与第一配送范围内盈亏相关的其它特征；
所述第二配送范围特征可以包括以下特征中的一个或多个：第二配送范围内用户的需求特征，包含用户喜好的特征、用户消费金额，与第二配送范围内盈亏相关的其它特征；
所述配送方特征可以包括以下特征中的一个或多个：配送方基本信息、配送方周边运力情况、与周围配送方对比构造的特征、第一配送范围与第二配送范围对比构造的特征。
步骤130：基于所述第二配送范围特征、第一配送范围特征和配送方特征，预测所述目标配送方采用混合送模式后所述第二配送范围内业务的变化值。
服务器在确定了目标配送方的第二配送范围特征，获取了所述第一配送范围的第一配送范围特征和所述目标配送方的配送方特征之后，就可以预测所述目标配送方采用混合送模式后所述第二配送范围内业务的变化值。
在一实施例中，服务器可以基于机器学习技术，预先训练用于预测所述目标配送方采用混合送模式后第二配送范围内业务变化值的机器学习模型。
即，所述步骤130，包括：
将所述第二配送范围特征、第一配送范围特征和配送方特征输入到预设的机器学习模型；
获取所述机器学习模型计算后输出的所述目标配送方采用混合送模式后所述第二配送范围内业务的变化值。
随着机器学习技术的发展，利用预先训练的机器学习模型预测目标配送方采用混合送模式后第二配送范围内业务的变化值，计算效率以及准确性较高。
以下介绍所述预设的机器学习模型的训练过程：
获取采用混合送模式的配送方历史产生的第二配送范围历史特征、第一配送范围历史特征和配送方历史特征以及第二配送范围内业务的历史变化值；
将所述第二配送范围历史特征、第一配送范围历史特征和配送方历史特征和以及第二配送范围内业务的历史变化值作为训练样本输入到机器学习算法进行训练；其中，所述第二配送范围历史特征、第一配送范围历史特征和配送方历史特征作为样本输入，所述第二配送范围内业务的历史变化值作为样本输出；
得到最终训练出的机器学习模型。
在一实施例中，服务器可以预先收集大量已经开通混合送模式的配送方历史产生的第二配送范围历史特征、第一配送范围历史特征和配送方历史特征，以及每个配送方在第二配送范围内业务的历史变化值。
然后，将每个配送方的第二配送范围历史特征、第一配送范围历史特征、配送方历史特征和历史变化值(成对数据)作为训练样本集(一般的可以以特征作为输入值，变化值作为输出值)，并基于机器学习算法，训练机器学习模型。通过持续地学习可以不断完善所述机器学习模型，当该机器学习模型达到跳出条件(例如针对所有配送方预测的预测变化值与实际变化值的准确度满足业务要求)时，即可以上线并使用机器学习模型。
在预测目标配送方业务的变化值时，将目标配送方的第二配送范围特征、第一配送范围特征和配送方特征作为输入数据输入到机器学习模型后，该机器学习模型就可以根据训练得到的算法计算并输出一个第二配送范围内业务的变化值。
所述机器学习模型可以采用例如逻辑回归(LR)，梯度提升决策树(GBDT)、随机森林算法等。
以随机森林算法为例，所述机器学习模型所采用的算法可以为：
例如，A＝f(B，C，D)
其中，A表示第二配送范围内业务的变化值，B表示第一配送范围特征，C表示第二配送范围特征，D表示配送方特征。
采用随机森林算法的机器学习模型，具有模型泛化性能较好、模型训练收敛快等优点。
步骤140：在所述第二配送范围内业务的变化值符合预设条件的情况下，确定所述目标配送方适合混合送模式。
所述预设条件可以是预先根据业务需求灵活设置的。
例如，所述预设条件还可以由目标配送方进行设置。这样，可以做到不同目标配送方具有不同的预设条件，以适应不同配送方的个性化需求，实现千人千面的效果。
在一实施例中，所述第二配送范围内业务的变化值可以包括目标配送方在第二配送区域内的盈利值。此时，所述预设条件可以与盈利值相关。
例如，在所述变化值即盈利值达到(等于或者大于)阈值的情况下，确定所述目标 配送方适合混合送模式。其中，所述阈值同样可以由目标配送方预先设置。
以商家为例，商家为了确保盈利，可以将阈值设置为正值；如此只有在第二配送范围内业务盈利的情况下，商家才允许开通混合送模式。
在一实施例中，商家为了提高知名度或者提高流量，还可以将阈值设为负值，即商家在开通混合送模式后允许第二配送范围内业务可以是亏损的；只要亏损金额不超过该阈值均可以开通混合送模式。
当然，以上所列举的仅是示例，实际应用中目标配送方可以基于业务需求灵活设置。
在一实施例中，所述第二配送范围内业务的变化值可以包括配送员取货时长，则预设条件可以为小于预设时长；即在所述配送员取货时长小于预设时长的情况下，确定所述目标配送方适合混合送模式。这样可以确保配送员取货时长不至于太久，避免配送超时。
在一实施例中，所述第二配送范围内业务的变化值可以包括物流接收方接收时长，则预设条件可以为小于预设时长；即在所述物流接收方接收时长小于预设时长的情况下，确定所述目标配送方适合混合送模式。这样可以确保物流接收方接收时长不至于太久，避免配送超时。
上述几种变化值可以组合使用，从而更为准确的确定目标配送方是否适合开通混合送模式。
在确定所述目标配送方适合混合送模式后，服务器可以为目标配送方开通混合送模式。
在一实施例中，在所述变化值符合预设条件的情况下，服务器可以为目标配送方开通混合送模式，将前述生成的目标配送方的第二配送范围生效，以使目标配送方的实际业务可以覆盖第二配送范围。例如，向第二配送范围内的用户推送目标配送方，以使第二配送范围内的用户可以看到目标配送方提供的服务。
综上所述，本申请提供了一种确定配送方适合混合送模式的方案，根据所述第二配送范围确定第二配送范围特征，以及获取所述第一配送范围的第一配送范围特征和所述目标配送方的配送方特征；根据所述第二配送范围特征、第一配送范围特征和配送方特征，预测所述目标配送方采用混合送模式后所述第二配送范围内业务的变化值；在所述第二配送范围内业务的变化值符合预设条件的情况下，确定所述目标配送方适合混合送模式。如此，通过预测配送方在采用混合送模式后第二配送范围内业务的变化情况判断 该配送方是否适合进行混合送，从而可以仅为适合进行混合送的配送方开通混合送模式。由于预测结果是基于客观存在的第一配送范围特征、第二配送范围特征以及配送方特征而确定的，因此预测结果也是客观符合自然规律的。
与前述确定配送方适合混合送模式的方法的实施例相对应，本申请还提供了确定配送方适合混合送模式的装置的实施例。
本申请确定配送方适合混合送模式的装置的实施例可以应用在服务器上。装置实施例可以通过软件实现，也可以通过硬件或者软硬件结合的方式实现。以软件实现为例，作为一个逻辑意义上的装置，是通过其所在处理器将非易失性存储器中对应的计算机程序指令读取到内存中运行形成的。从硬件层面而言，如图3所示，为本申请识别配送方适合混合送模式的装置所在的一种硬件结构图，除了图3所示的处理器、内存、网络接口、以及非易失性存储器之外，实施例中通常根据该确定配送方适合混合送模式的实际功能，还可以包括其他硬件，对此不再赘述。
请参考图4，在一实施例中，该确定配送方适合混合送模式的装置可以包括：
生成单元310，基于目标配送方的第一配送范围生成所述目标配送方的第二配送范围；其中，所述第二配送范围包围所述第一配送范围，所述目标配送方未开通混合送模式且有意向开通混合送模式；
特征获取单元320，基于所述第二配送范围确定第二配送范围特征，以及获取所述第一配送范围的第一配送范围特征和所述目标配送方的配送方特征；
预测单元330，基于所述第二配送范围特征、第一配送范围特征和配送方特征，预测所述目标配送方采用混合送模式后所述第二配送范围内业务的变化值；
确定单元340，在所述第二配送范围内业务的变化值符合预设条件的情况下，确定所述目标配送方适合混合送模式。
可选的，所述生成单元310，包括：
以第一配送范围的边界向外扩展预设距离，将扩展预设距离的区域确定为第二配送范围。
可选的，所述生成单元310，包括：
扩展子单元，以构建第一配送范围的边界坐标点为基准点，每个所述基准点分别向外扩展预设距离得到第二配送范围的边界坐标点；
生成子单元，根据多个第二配送范围的边界坐标点中相邻的边界坐标点之间的导航路径，确定第二配送范围的边界；其中，所述第二配送范围的边界与第一配送范围的边界之间的区域为第二配送范围。
可选的，所述预测单元330，包括：
输入子单元，将所述第二配送范围特征、第一配送范围特征和配送方特征输入到预设的机器学习模型；
输出子单元，获取所述机器学习模型计算后输出的所述目标配送方采用混合送模式后第二配送范围内业务的变化值。
可选的，所述装置还包括，对所述预设的机器学习模型采用以下子单元进行训练：
获取子单元，获取采用混合送模式的配送方历史产生的第二配送范围历史特征、第一配送范围历史特征和配送方历史特征以及第二配送范围内业务的历史变化值；
训练子单元，将所述第二配送范围历史特征、第一配送范围历史特征和配送方历史特征和第二配送范围内业务的历史变化值作为训练样本输入到机器学习算法训练；其中，所述第二配送范围历史特征、第一配送范围历史特征和配送方历史特征作为样本输入，所述第二配送范围业务的历史变化值作为样本输出；
确定子单元，得到最终训练出的机器学习模型。
可选的，所述机器学习算法包括随机森林算法。
可选的，所述确定单元340，包括：
在所述变化值达到阈值的情况下，确定所述目标配送方适合混合送模式。
可选的，所述第一配送范围特征，包括：
配送方在第一配送范围内的历史订单分布、历史的配送体验或历史订单收益中的至少一个；
所述第二配送范围特性，包括：
第二配送范围内用户喜好或用户消费金额中的至少一个；
所述配送方特征，包括：
配送方基本信息或配送方周边运力中的至少一个。
上述装置中各个单元的功能和作用的实现过程具体详见上述方法中对应步骤的实现过程，在此不再赘述。
对于装置实施例而言，由于其基本对应于方法实施例，所以相关之处参见方法实施例的部分说明即可。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申请方案的目的。本领域普通技术人员在不付出创造性劳动的情况下，即可以理解并实施。
以上图4描述了识别配送方适合混合送模式的装置的内部功能模块和结构示意图，其实质上的执行主体可以为一种电子设备，包括：
处理器；
用于存储处理器可执行指令的存储器；
其中，所述处理器被配置为：
基于目标配送方的第一配送范围生成所述目标配送方的第二配送范围；其中，所述第二配送范围包围所述第一配送范围，所述目标配送方未开通混合送模式且有意向开通混合送模式；
基于所述第二配送范围确定第二配送范围特征，以及获取所述第一配送范围的第一配送范围特征和所述目标配送方的配送方特征；
基于所述第二配送范围特征、第一配送范围特征和配送方特征，预测所述目标配送方采用混合送模式后所述第二配送范围内业务的变化值；
在所述第二配送范围内业务的变化值符合预设条件的情况下，确定所述目标配送方适合混合送模式。
可选的，所述基于目标配送方的第一配送范围生成所述目标配送方的第二配送范围，包括：
以第一配送范围的边界向外扩展预设距离，将扩展预设距离的区域确定为第二配送范围。
可选的，所述以第一配送范围的边界向外扩展预设距离，将扩展预设距离的区域确定为第二配送范围，包括：
以构建第一配送范围的边界坐标点为基准点，每个基准点分别向外扩展预设距离得到第二配送范围的边界坐标点；
根据多个第二配送范围的边界坐标点中相邻的边界坐标点之间的导航路径，确定第二配送范围的边界；其中，所述第二配送范围的边界与第一配送范围的边界之间的区域为第二配送范围。
可选的，所述基于所述第二配送范围特征、第一配送范围特征和配送方特征，预测所述目标配送方采用混合送模式后所述第二配送范围内业务的变化值，包括：
将所述第二配送范围特征、第一配送范围特征和配送方特征输入到预设的机器学习模型；
获取所述机器学习模型计算后输出的所述目标配送方采用混合送模式后所述第二配送范围内业务的变化值。
可选的，还包括，对所述预设的机器学习模型采用以下方式进行训练：
获取采用混合送模式的配送方历史产生的第二配送范围历史特征、第一配送范围历史特征和配送方历史特征以及第二配送范围内业务的历史变化值；
将所述第二配送范围历史特征、第一配送范围历史特征和配送方历史特征和所述第二配送范围内业务的历史变化值作为训练样本输入到机器学习算法训练；其中，所述第二配送范围历史特征、第一配送范围历史特征和配送方历史特征作为样本输入，所述第二配送范围内业务的历史变化值作为样本输出；
得到最终训练出的机器学习模型。
可选的，所述机器学习算法包括随机森林算法。
可选的，所述在所述第二配送范围内业务的变化值符合预设条件的情况下，确定所述目标配送方适合混合送模式，包括：
在所述第二配送范围内业务的变化值达到阈值的情况下，确定所述目标配送方适合混合送模式。
可选的，所述第一配送范围特征，包括：
配送方在第一配送范围内的历史订单分布、历史的配送体验或历史订单收益中的至少一个；
所述第二配送范围特性，包括：
第二配送范围内用户喜好或用户消费金额中的至少一个；
所述配送方特征，包括：
配送方基本信息或配送方周边运力中的至少一个。
在上述电子设备的实施例中，应理解，该处理器可以是中央处理单元(英文：Central Processing Unit，简称：CPU)，还可以是其他通用处理器、数字信号处理器(英文：Digital Signal Processor，简称：DSP)、专用集成电路(英文：Application Specific Integrated Circuit，简称：ASIC)等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等，而前述的存储器可以是只读存储器(英文：read-only memory，缩写：ROM)、随机存取存储器(英文：random access memory，简称：RAM)、快闪存储器、硬盘或者固态硬盘。结合本发明实施例所公开的方法的步骤可以直接体现为硬件处理器执行完成，或者用处理器中的硬件及软件模块组合执行完成。
本申请中的各个实施例均采用递进的方式描述，各个实施例之间相同相似的部分互相参见即可，每个实施例重点说明的都是与其他实施例的不同之处。尤其，对于电子设备实施例而言，由于其基本相似于方法实施例，所以描述的比较简单，相关之处参见方法实施例的部分说明即可。
以上所述仅为本申请的较佳实施例而已，并不用以限制本申请，凡在本申请的精神和原则之内，所做的任何修改、等同替换、改进等，均应包含在本申请保护的范围之内。

Claims (11)

  1. 一种确定配送方适合混合送模式的方法，包括：
    基于目标配送方的第一配送范围生成所述目标配送方的第二配送范围；其中，所述第二配送范围包围所述第一配送范围，所述目标配送方暂未开通所述混合送模式且有意向开通所述混合送模式；
    基于所述第二配送范围确定第二配送范围特征，以及获取所述第一配送范围的第一配送范围特征和所述目标配送方的配送方特征；
    基于所述第二配送范围特征、所述第一配送范围特征和所述配送方特征，预测所述目标配送方采用所述混合送模式后所述第二配送范围内业务的变化值；
    在所述第二配送范围内业务的变化值符合预设条件的情况下，确定所述目标配送方适合混合送模式。
  2. 根据权利要求1所述的方法，其特征在于，基于所述目标配送方的第一配送范围生成所述目标配送方的第二配送范围，包括：
    以所述第一配送范围的边界向外扩展预设距离，将所述扩展预设距离的区域确定为所述第二配送范围。
  3. 根据权利要求2所述的方法，其特征在于，以所述第一配送范围的边界向外扩展所述预设距离，将所述扩展预设距离的区域确定为所述第二配送范围，包括：
    以构建所述第一配送范围的边界坐标点为基准点，将每个所述基准点分别向外扩展所述预设距离得到所述第二配送范围的边界坐标点；
    根据多个所述第二配送范围的边界坐标点中相邻的边界坐标点之间的导航路径，确定所述第二配送范围的边界；其中，所述第二配送范围的边界与所述第一配送范围的边界之间的区域为所述第二配送范围。
  4. 根据权利要求1-3中任一项所述的方法，其特征在于，基于所述第二配送范围特征、所述第一配送范围特征和配送方特征，预测所述目标配送方采用混合送模式后所述第二配送范围内业务的变化值，包括：
    将所述第二配送范围特征、所述第一配送范围特征和所述配送方特征输入到预设的机器学习模型；
    获取所述机器学习模型计算后输出的所述目标配送方采用混合送模式后所述第二 配送范围内业务的变化值。
  5. 根据权利要求4所述的方法，还包括，对所述预设的机器学习模型采用以下方式进行训练：
    获取采用所述混合送模式的配送方历史产生的第二配送范围历史特征、第一配送范围历史特征和配送方历史特征以及第二配送范围内业务的历史变化值；
    将所述第二配送范围历史特征、所述第一配送范围历史特征和所述配送方历史特征和所述第二配送范围内业务的历史变化值作为训练样本输入到机器学习算法进行训练；其中，所述第二配送范围历史特征、所述第一配送范围历史特征和所述配送方历史特征作为样本输入，所述第二配送范围内业务的历史变化值作为样本输出；
    得到最终训练出的所述机器学习模型。
  6. 根据权利要求5所述的方法，其特征在于，所述机器学习算法包括随机森林算法。
  7. 根据权利要求1-6中任一项所述的方法，其特征在于，在所述第二配送范围内业务的变化值符合所述预设条件的情况下，确定所述目标配送方适合所述混合送模式，包括：
    在所述第二配送范围内业务的变化值达到阈值的情况下，确定所述目标配送方适合所述混合送模式。
  8. 根据权利要求1-7中任一项所述的方法，其特征在于，
    所述第一配送范围特征包括：
    配送方在所述第一配送范围内的历史订单分布、历史的配送体验或历史订单的收益中的至少一个；
    所述第二配送范围特性包括：
    所述第二配送范围内用户喜好或用户消费金额中的至少一个；
    所述配送方特征包括：
    配送方基本信息或配送方周边运力中的至少一个。
  9. 一种确定配送方适合混合送模式的装置，包括：
    生成单元，基于目标配送方的第一配送范围生成所述目标配送方的第二配送范围； 其中，所述第二配送范围包围所述第一配送范围，所述目标配送方未开通所述混合送模式且有意向开通所述混合送模式；
    特征获取单元，基于所述第二配送范围确定第二配送范围特征，以及获取所述第一配送范围的第一配送范围特征和所述目标配送方的配送方特征；
    预测单元，基于所述第二配送范围特征、所述第一配送范围特征和所述配送方特征，预测所述目标配送方采用所述混合送模式后所述第二配送范围内业务的变化值；
    确定单元，在所述第二配送范围内业务的变化值符合预设条件的情况下，确定所述目标配送方适合混合送模式。
  10. 一种计算机可读存储介质，存储有计算机程序，所述计算机程序用于执行上述权利要求1-8中任一项所述的方法。
  11. 一种电子设备，包括：
    处理器；
    用于存储处理器可执行指令的存储器；
    所述处理器被配置为执行上述权利要求1-8中任一项所述的方法。
PCT/CN2019/125168 2019-06-25 2019-12-13 配送方适合混合送模式的确定 Ceased WO2020258756A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910557250.8A CN110288296B (zh) 2019-06-25 2019-06-25 识别配送方适合混合送模式的方法及装置
CN201910557250.8 2019-06-25

Publications (1)

Publication Number Publication Date
WO2020258756A1 true WO2020258756A1 (zh) 2020-12-30

Family

ID=68005655

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/125168 Ceased WO2020258756A1 (zh) 2019-06-25 2019-12-13 配送方适合混合送模式的确定

Country Status (2)

Country Link
CN (1) CN110288296B (zh)
WO (1) WO2020258756A1 (zh)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110288296B (zh) * 2019-06-25 2025-01-28 北京三快在线科技有限公司 识别配送方适合混合送模式的方法及装置
CN112348424B (zh) * 2020-05-08 2024-07-19 北京京东振世信息技术有限公司 一种末端包裹配送方法和装置
CN111967945A (zh) * 2020-08-28 2020-11-20 上加下信息技术成都有限公司 一种订单配送方法、装置及电子设备
CN114936819B (zh) * 2022-05-12 2024-12-20 拉扎斯网络科技（上海）有限公司 配送范围生成方法及装置
CN116611677B (zh) * 2023-07-21 2024-01-19 广州市大枣信息科技有限公司 一种外卖订单推荐方法及系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107093038A (zh) * 2016-11-18 2017-08-25 北京小度信息科技有限公司 配送方式选择方法和装置
CN109598430A (zh) * 2018-11-28 2019-04-09 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
CN109615122A (zh) * 2018-11-27 2019-04-12 拉扎斯网络科技（上海）有限公司 配送范围的生成方法、装置、电子设备和存储介质
CN110288296A (zh) * 2019-06-25 2019-09-27 北京三快在线科技有限公司 识别配送方适合混合送模式的方法及装置

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010035498A1 (ja) * 2008-09-29 2010-04-01 日本たばこ産業株式会社 配送方式決定支援システム、コース作成システム、配送方式決定支援方法、およびコース作成方法
US10176448B1 (en) * 2015-12-30 2019-01-08 Square, Inc. Generation of dynamic delivery zones for merchants
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
CN107437122B (zh) * 2016-05-26 2019-03-26 北京三快在线科技有限公司 确定商家配送范围的方法及装置
CN109426896A (zh) * 2017-08-23 2019-03-05 北京三快在线科技有限公司 一种确定区域运力状态的方法及装置、电子设备
CN109636013B (zh) * 2018-11-27 2021-06-29 拉扎斯网络科技（上海）有限公司 配送范围的生成方法、装置、电子设备和存储介质

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107093038A (zh) * 2016-11-18 2017-08-25 北京小度信息科技有限公司 配送方式选择方法和装置
CN109615122A (zh) * 2018-11-27 2019-04-12 拉扎斯网络科技（上海）有限公司 配送范围的生成方法、装置、电子设备和存储介质
CN109598430A (zh) * 2018-11-28 2019-04-09 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
CN110288296A (zh) * 2019-06-25 2019-09-27 北京三快在线科技有限公司 识别配送方适合混合送模式的方法及装置

Also Published As

Publication number Publication date
CN110288296B (zh) 2025-01-28
CN110288296A (zh) 2019-09-27

Similar Documents

Publication Publication Date Title
WO2020258756A1 (zh) 配送方适合混合送模式的确定
US20230169448A1 (en) Delivery prediction generation system
US12298759B2 (en) Using social media data of a vehicle occupant to alter a route plan of the vehicle
US12166835B2 (en) Push notification delivery system with feedback analysis
US20240410712A1 (en) Optimization of network service based on an existing service
US10977751B1 (en) Managing communications for combined orders
US20180285827A1 (en) Distinguishing events of users for efficient service content distribution
US11475218B2 (en) Apparatus and method for providing sentence based on user input
US11580549B2 (en) Transaction tracking and fraud detection using voice and/or video data
US20160164974A1 (en) Service Content Tailored To Out Of Routine Events
WO2018006469A1 (zh) 基于知识图谱的人机交互方法及系统
CN107683486A (zh) 用户事件的具有个人影响性的改变
US20200160386A1 (en) Control of advertisement delivery based on user sentiment
US10542105B2 (en) Geo-locating individuals based on a derived social network
US20100250366A1 (en) Merge real-world and virtual markers
US20250348518A1 (en) Systems, devices, and methods for content selection
US10366134B2 (en) Taxonomy-based system for discovering and annotating geofences from geo-referenced data
US9460348B1 (en) Associating location history with photos
Huq et al. Profit and satisfaction aware order assignment for online food delivery systems exploiting water wave optimization
JP2017059032A (ja) 推定装置
US12101692B2 (en) Wireless tracking and data queue management system
JP7138209B1 (ja) 受付装置、受付方法、受付プログラム
CN115481961A (zh) 一种预测配送资源到达商户时间的方法及装置

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: 19935579

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: 19935579

Country of ref document: EP

Kind code of ref document: A1