CN109636013B

CN109636013B - Distribution range generation method and device, electronic equipment and storage medium

Info

Publication number: CN109636013B
Application number: CN201811426898.3A
Authority: CN
Inventors: 叶莺; 李淳敏
Original assignee: Rajax Network Technology Co Ltd
Current assignee: Rajax Network Technology Co Ltd
Priority date: 2018-11-27
Filing date: 2018-11-27
Publication date: 2021-06-29
Anticipated expiration: 2038-11-27
Also published as: CN109636013A

Abstract

The embodiment of the invention relates to the technical field of communication, and discloses a method and a device for generating a distribution range, electronic equipment and a storage medium, wherein the method comprises the steps of acquiring order characteristic data of each distribution unit in a first distribution range of a target object; obtaining target delivery units from each delivery unit according to the order characteristic data and a preset first threshold value; the target delivery unit is deleted within the first delivery range. According to the invention, the more reasonable distribution range is determined by deleting the areas with lower distribution efficiency and poor distribution quality, so that the distribution quality and the distribution efficiency are improved.

Description

Distribution range generation method and device, electronic equipment and storage medium

Technical Field

The present invention relates to the field of communications technologies, and in particular, to a method and an apparatus for generating a distribution range, an electronic device, and a storage medium.

Background

The current way to determine the distribution scope of the merchant is generally as follows: the distribution range is generated by taking the merchant as the center and taking the distribution distance or the distribution time length as the radius. However, the distribution range obtained by the existing method for determining the distribution range according to the distribution distance can cause that some areas with difficult distribution or poor distribution quality of orders exist in the merchant; according to the distribution range obtained by the existing method for determining the distribution range by the distribution time length, the time length circle is static and cannot be changed randomly, and the riding speed is a dynamic change process due to the constraint of traffic conditions, weather conditions and rider conditions, so that the formed distribution range is not optimized enough, and the order distribution efficiency is influenced.

Disclosure of Invention

The invention aims to provide a method and a device for generating a distribution range, an electronic device and a storage medium, which are used for optimizing the distribution range of a merchant and improving the distribution quality by eliminating areas with low distribution efficiency and poor distribution quality and invalid areas where logistics such as rivers, lakes, seas, scenic spots, parks, cemeteries and the like cannot be distributed.

In order to solve the above technical problem, an embodiment of the present invention provides a method for generating a delivery range, including: the method comprises the steps of obtaining order characteristic data of each delivery unit in a first delivery range of a target object; obtaining target delivery units from each delivery unit according to the order characteristic data and a preset first threshold value; the target delivery unit is deleted within the first delivery range.

An embodiment of the present invention further provides a distribution range generation apparatus, including: the data acquisition module is used for acquiring order characteristic data of each distribution unit in a first distribution range of the target object; the grid judgment module is used for obtaining target delivery units from the delivery units according to the order characteristic data and a preset first threshold value; and the grid filtering module deletes the target delivery unit in the first delivery range.

Embodiments of the present invention also provide an electronic device, comprising at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to implement: the method comprises the steps of obtaining order characteristic data of each delivery unit in a first delivery range of a target object; obtaining target delivery units from each delivery unit according to the order characteristic data and a preset first threshold value; the target delivery unit is deleted within the first delivery range.

Embodiments of the present invention also provide a nonvolatile storage medium storing a computer-readable program for causing a computer to execute the method of generating a delivery range as described above.

Compared with the prior art, the implementation mode of the invention has the main differences and the effects that: according to the invention, the delivery units which do not meet the requirements are deleted in the initial first delivery range of the target object, namely, some areas with lower delivery efficiency and poor delivery quality in the existing delivery range are removed, so that the delivery range of a merchant is optimized, and the delivery range is more reasonable.

In addition, the first threshold includes an effective completion unit quantile, the order characteristic data includes an effective completion unit quantity, and the target delivery unit is obtained from the delivery unit according to the order characteristic data and the preset first threshold, specifically: judging whether the effective completion unit quantity of the distribution unit is smaller than the effective completion unit quantile or not; and if the effective completion unit quantity of the delivery unit is less than the effective completion unit quantile, determining the delivery unit as the target delivery unit. When the effective completion single amount of a certain delivery unit is less, the delivery efficiency of the sub-area unit is considered to be lower, and the delivery range of the target object can be optimized by removing the delivery unit from the delivery range.

In addition, the effective completion monodispersion is obtained according to the following method: determining a distribution range set of all objects in a historical time period by a distribution site corresponding to a target object; sequencing all the distribution units in the distribution range set according to the size of the effective completion single quantity; and determining a corresponding reference distribution unit in the sequence according to a preset percentage reference, and taking the effective completion single quantity of the reference distribution unit as an effective completion single quantile.

In addition, the first threshold includes an overtime single proportion reference, the order characteristic data includes an overtime single quantity and an effective finished single quantity, and the target delivery unit is obtained from the delivery unit according to the order characteristic data and the preset first threshold, specifically: judging whether the ratio of the overtime single quantity to the effective completion single quantity of the distribution unit is larger than an overtime ratio reference or not; and if the ratio of the distribution units is larger than the overtime ratio reference, determining the distribution unit as a target distribution unit. When the overtime probability of a certain distribution unit is higher, the distribution quality of the distribution unit is considered to be poor, and the rationality of the distribution range of the merchant can be improved by removing the sub-area unit.

In addition, the timeout duty cycle reference is obtained according to the following method: counting effective finished orders and overtime orders of all objects in a historical time period of a delivery site corresponding to a target object; and taking the ratio of the overtime order quantity of all orders to the effective finished order quantity as the overtime order proportion reference.

In addition, the first threshold includes a standard exceeding a preset time-length unit ratio, the order characteristic data includes a unit quantity exceeding the preset time-length and an effective unit quantity, and the target delivery unit is obtained from the delivery unit according to the order characteristic data and the preset first threshold, specifically: judging whether the ratio of the single quantity exceeding the preset time length of the distribution unit to the single quantity effectively completed is larger than the reference of the ratio exceeding the preset time length of the single quantity; and if the ratio of the distribution units is larger than the preset time-length single-occupation ratio reference, determining the distribution units as target distribution units. When the overtime probability in the preset duration of a certain distribution unit is larger, the distribution quality of the sub-area unit is considered to be poor, and the rationality of the distribution range of the merchant can be further improved by removing the sub-area unit.

In addition, the single proportion reference exceeding the preset time length is obtained according to the following method: counting effective completion orders and units exceeding preset time duration of all the orders of all the objects in a historical time period of a distribution site corresponding to the target object; and taking the ratio of the units with the exceeding preset time length of all orders to the effective completion units as the reference of the ratio of the units with the exceeding preset time length.

In addition, in order to compensate for the reduced number of delivery units, the method for generating a delivery range further includes: acquiring a target area corresponding to a target object, and determining a plurality of sub-area units in the target area, wherein the target area is larger than a first distribution range, and the plurality of sub-area units are positioned in the target area outside the first distribution range; obtaining an expanded sub-region unit according to the attribute information and the position information of the target object, the position information of the sub-region unit and a preset distribution prediction model; and adding the expanded sub-area unit into a second distribution range of the target object, wherein the second distribution range is the distribution range generated by deleting the target distribution unit in the first distribution range. After unreasonable areas with low delivery efficiency and poor delivery quality are removed, in order to make up for the reduced order quantity, the areas are expanded, the delivery range is further expanded, and the more optimal delivery range is finally determined.

In addition, obtaining an expanded sub-region unit according to the attribute information, the position information and the position information of the target object and a preset delivery prediction model, specifically: determining the predicted order quantity of the sub-area unit according to the attribute information of the target object and a preset first delivery prediction model; when the predicted order quantity is larger than the preset order quantity reference, determining the predicted delivery duration of the sub-area unit according to the position information of the target object, the position information of the sub-area unit and a preset second delivery prediction model; and when the predicted distribution time length is less than the average distribution time length, taking the sub-area unit as an expanded sub-area unit. Firstly, whether a certain sub-area unit can generate an order is judged, when the order can be generated is predicted, the distribution time length of the sub-area unit is judged, if the distribution time length is shorter, the distribution quality and the distribution efficiency of the sub-area unit are higher, and the sub-area unit can be used as an expanded area to be added into the distribution range of a target object.

In addition, the method further comprises: constructing a first delivery prediction model, wherein the first delivery prediction model comprises a first prediction function, and the first prediction function takes attribute information of an object as a parameter; according to a machine learning algorithm, training a first delivery prediction model by using order characteristic data of all orders of a city where a target object is located in a historical time period; determining a predicted order quantity of a certain sub-area unit according to the attribute information of the target object and a preset first delivery prediction model, specifically comprising: and inputting the attribute information of the target object in the trained first delivery prediction model to obtain the predicted order quantity of the sub-region unit. The most accurate predicted quantity is obtained by intelligently training the order quantity, and when the predicted order quantity of a certain sub-area unit is large, the sub-area unit can be added into a distribution range.

In addition, the method further comprises: constructing a second delivery prediction model, wherein the second delivery prediction model comprises a second prediction function, and the second prediction function takes the position information of the object and the position information of the sub-area unit as parameters; according to a machine learning algorithm, training a second delivery prediction model by using delivery duration data of all orders in a historical time period of a city where a target object is located; determining the predicted delivery duration of the sub-area unit according to the position information of the target object, the position information of the sub-area unit and a preset second delivery prediction model, and specifically comprising the following steps: and inputting the position information of the target object and the position information of a certain sub-area unit in the trained second distribution prediction model to obtain the predicted distribution time length of the sub-area unit. And obtaining the most accurate predicted time length according to the intelligent training, and adding a sub-region unit into a distribution range when the predicted time length of order distribution of the sub-region unit is smaller.

In addition, the order characteristic data of each sub-area unit in the first area of the target object is obtained, and the method comprises the steps of extracting order data of a first distribution range from a database; and screening the order data to obtain order characteristic data of each distribution unit. The order characteristic data can provide characteristic data with the most influence, and unreasonable distribution units can be eliminated conveniently through calculation.

In addition, when generating the delivery range, the method further includes: acquiring a geo-fence comprising preset keywords; judging each distribution unit and sub-area unit in the second distribution range according to the geo-fence to obtain an invalid unit; the invalid unit is deleted in the second delivery range. By deleting invalid areas where logistics such as rivers, lakes and seas, scenic spots, parks, cemeteries and the like cannot be distributed, the order distribution range is further optimized.

Drawings

Fig. 1 is a flowchart of a method of generating a delivery range according to a first embodiment of the present invention;

FIG. 2-1 is a flowchart of a method for generating a delivery range according to a second embodiment of the present invention;

FIG. 2-2 is a flowchart of an effective completion single quantile acquisition method according to a second embodiment of the present invention;

FIG. 3-1 is a flowchart of a method for generating a delivery range according to a third embodiment of the present invention;

fig. 3-2 is a flowchart of a timeout single proportion reference acquisition method according to a third embodiment of the present invention;

FIG. 4-1 is a flowchart of a method for generating a delivery range according to a fourth embodiment of the present invention;

fig. 4-2 is a flowchart of a timeout single-proportion reference obtaining method within a preset duration according to a fourth embodiment of the present invention;

FIG. 5-1 is a flowchart of a method of generating a delivery range according to a fifth embodiment of the present invention;

FIG. 5-2 is a schematic view of a method for obtaining the expansion sub-region unit according to the fifth embodiment of the present invention;

fig. 6 is a flowchart of a method of generating a delivery range according to a sixth embodiment of the present invention;

FIG. 7-1 is a flowchart of a delivery range generating method according to a seventh embodiment of the present invention;

FIG. 7-2 is a schematic view of a first delivery range and a target area according to a seventh embodiment of the present invention;

fig. 7-3 are schematic diagrams of intelligent delivery ranges generated by the delivery range generation method according to the seventh embodiment of the present invention;

fig. 8 is a schematic diagram of a distribution range generating apparatus according to an eighth embodiment of the present invention;

fig. 9 is a schematic structural diagram of an electronic device according to a ninth embodiment of the invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, embodiments of the present invention will be described in detail below with reference to the accompanying drawings. However, it will be appreciated by those of ordinary skill in the art that numerous technical details are set forth in order to provide a better understanding of the present application in various embodiments of the present invention. However, the technical solution claimed in the present application can be implemented without these technical details and various changes and modifications based on the following embodiments.

The first embodiment of the present invention relates to a method for generating a distribution range, and the present invention may be applied to a terminal side, such as a terminal device such as a mobile phone and a tablet computer, and may also be applied to a server on a network side. As shown in fig. 1, the method for generating a distribution range according to the present embodiment includes:

step 101, obtaining order characteristic data of each delivery unit in a first delivery range of a target object;

102, obtaining target delivery units from each delivery unit according to order characteristic data and a preset first threshold value;

step 103, delete the target delivery unit within the first delivery range.

Specifically, the target object may be a merchant restaurant, a confectionary shop, a pharmacy store, a distribution site, or the like, for example, a geographical range is formed by radiating a certain distance outwards with the longitude and latitude of a certain merchant as a center, and the geographical range is screened, for example, a regularly or irregularly shaped geographical range formed after removing an area, such as a highway, a bridge, or the like, which is not possible to be an order arrival point is used as an initial distribution range (i.e., a first distribution range) of the merchant. The initial distribution range is then divided into a plurality of distribution units (each distribution unit may be a shape obtained by equally dividing the first area, or may be obtained by dividing the first area according to a certain rule, such as dividing the first area according to a geographical location), and it should be specifically noted that each distribution unit may be a grid, for example, a square grid of 140 meters × 140 meters, or a unit grid of other shapes and sizes.

In step 101, order characteristic data is obtained according to the following method: extracting order data of a first distribution range from a database; and screening the order data to obtain order characteristic data of each distribution unit.

Explaining by taking a target object as a merchant, collecting the order data related to the merchant in the past in a database, and extracting order distribution data of a first area in the database, wherein the order data comprises an invoice broad table (a database table consisting of merchant coordinates, meal taking time, meal delivery time and the like, the broad table refers to a database table with more fields, and generally refers to a database table with related indexes, dimensions and attributes related to business topics), an order broad table (a database table consisting of restaurant categories, meal categories and the like of historical orders), a merchant broad table (a database table consisting of meal categories, merchant coordinates, merchant labels and the like), and a POI interface of an electronic map (an Apache POI is an open source item of an Apache software foundation, and the POI provides an API for Java program to read and write a Microsoft Office format file); and after the order data are sorted, summarized and screened, the order characteristic data of each distribution unit in the first distribution range are obtained. The order characteristic data may be characteristic data that has an effect on the delivery quality and efficiency of the order, facilitating subsequent calculations based on the order characteristic data to optimize the existing delivery scope.

In practical applications, the order characteristic data obtained after processing the order distribution data such as the waybill width table, the order width table, the merchant width table, the POI interface of the electronic map, and the like may include a merchant ID, an ID of a distribution unit where the waybill is located, a walking distance from the merchant to the distribution unit, an average distribution time, an average AOI meal delivery time (a time when a rider arrives at a destination and goes upstairs or enters a cell to find a corresponding resident), an average meal delivery time, an effective completion order, a timeout order within a preset time, and the like. Taking the take-away industry as an example, wherein the effective completion order refers to an order form in which the rider successfully delivers the meal to the user; the overtime order is an order form of which the food delivery is overtime due to the fact that the rider does not deliver the food to the user within the specified time; the order that the rider sends the food overtime within the preset duration is the order that the rider sends the food overtime within the preset duration.

Further, the order distribution data is screened to obtain order characteristic data of each distribution unit in the first distribution range, specifically, an order of a certain target merchant falling into each distribution unit (grid) is screened out through a merchant ID in the order distribution data and an ID of a distribution unit where the freight note is located, and then related order data is screened as the order characteristic data according to needs.

In this embodiment, after the order characteristic data (e.g., effective finished order quantity) of each delivery unit in the first delivery range is acquired in step 102, the order characteristic data needs to be compared with a preset first threshold, and the principle of the comparison is to use the delivery unit corresponding to the order characteristic data meeting the first threshold as the target delivery unit.

For how to determine the distribution range of a merchant, the existing distribution range calculation schemes include two, namely a time length circle based on time length calculation and a riding circle based on path planning calculation. Time-length circle: the delivery duration is divided into three segments in total, namely, the fetching time of a rider (or called the meal-out duration), the riding duration of the rider (ETS), the duration from the arrival of the rider at the AOI of the user (the arrival of the rider at the destination to go upstairs or the arrival of the rider at a cell to find a corresponding resident) (ETA), and the duration circle of the delivery range (the duration circle established by the riding duration ETS of the rider). Specifically, the method for determining the distribution range of the businessman based on the time-length circle is that the riding time length is sampled outwards from the center point of the businessman, and a group of sampling time length points with discrete periphery are connected with each other to form the time-length circle. And (4) riding a ring: and (4) utilizing path planning, inputting the riding length radius after data analysis, and generating a riding ring by taking the radius as the basis of the path planning.

Different from the existing distribution range calculation scheme, in the embodiment, a large geographical range is divided into each small distribution unit, target distribution units are obtained by obtaining order characteristic data of each distribution unit and judging each distribution unit in a first distribution range according to the order characteristic data and a preset first threshold, the target distribution units are areas with low distribution efficiency and poor distribution quality, and the target distribution units are deleted in the first distribution range, so that a more reasonable distribution range is determined.

A second embodiment of the present invention provides a method for generating a delivery range, as shown in fig. 2-1, including:

step 201, obtaining order characteristic data of each delivery unit in a first delivery range of a target object;

step 202, judging whether the effective completion unit quantity of the distribution unit is smaller than the effective completion unit quantile;

step 203, if the effective completion unit quantity of the distribution unit is less than the effective completion unit quantile, determining the distribution unit as a target distribution unit;

in step 204, the target delivery unit is deleted from the first delivery range, and a second delivery range of the target object is generated.

Step 201 and step 204 of this embodiment are the same as step 101 and step 103 of the first embodiment, and are not described again here. Step 202 and step 203 are described in detail below.

In this embodiment, the first threshold includes an effective completion single quantile, the order characteristic data includes an effective completion single quantity, the effective completion single quantity is an order quantity of the rider successfully delivering the meal to the user, and as shown in fig. 2-2, the effective completion single quantile is obtained by:

step 2021, determining a distribution range set of all the objects in a historical time period of the distribution site corresponding to the target object;

step 2022, sorting all the distribution units in the distribution range set according to the size of the effective completion single quantity;

step 2023, determining the corresponding reference distribution unit in the sequence according to the preset percentage reference, and using the effective completion unit quantity of the reference distribution unit as the effective completion unit quantile.

Specifically, in the takeaway delivery process, a takeaway order of a certain merchant is distributed by a delivery site, wherein the delivery site is a site which performs a delivery process for a logistics downstream dealer, a retailer and a customer in a certain business circle (such as a quiet temple business circle and a pentagon business circle) in a logistics supply chain link, and the delivery site utilizes circulation facilities and an information system platform to perform inversion, classification, circulation processing, matching and design a transportation route and a transportation mode on goods of a logistics manager, so that the distribution service is provided for a service object. At least one distribution site is usually set up in a certain business circle, wherein one distribution site can correspond to a plurality of merchants so as to carry out distribution procedures for orders of the merchants. When the effective completion unit quantile is determined, firstly determining a distribution site corresponding to the certain merchant, and then counting a distribution range set of all order merchants accepted by the distribution site in a historical time period, wherein the distribution range set can be a union of distribution ranges of all merchants, and the distribution range set can also be decomposed into a plurality of distribution units, and each distribution unit has order data in the historical time period; extracting effective completion single quantities of all the delivery units in the historical time period from the order data, and sequencing the delivery units according to the size of the effective completion single quantities, for example, sequencing the delivery units from large to small according to the effective completion single quantities; if a percentage reference is preset (for example, N% is preset), the last delivery unit sorted in the top N% is intercepted as a reference delivery unit, and the effective completion unit amount corresponding to the reference delivery unit is taken as an effective completion unit quantile.

For example, the takeaway orders of the merchant a are distributed by the delivery point W, first, the order delivery range set of all 50 merchants accepted by the site W in the past 10 days is counted, the delivery range set includes 300 delivery units W1, W2 · W300, the 300 delivery units are sorted from large to small according to the effective completion unit quantity, the preset percentage standard is 20%, the last delivery unit W69 sorted in the top 20% is intercepted as the standard delivery unit, the effective completion unit quantity (for example, 200 units) of the standard delivery unit is taken as the effective completion unit quantile, and then 200 units are taken as the effective completion unit quantiles.

According to the method provided by the embodiment, when the delivery units are determined according to the order characteristic data and the first threshold, taking the merchant a as an example, the first delivery range is assumed to be composed of 123 delivery units, processing the distribution data (such as the waybill width table, the order width table, the merchant width table, the POI interface of the electronic map, etc.) of the 123 distribution units to obtain the order characteristic data (such as the effective completion order data) of each distribution unit, wherein the effective completion list data of the delivery unit A1, the delivery unit A2 and the delivery unit A3 are respectively 160, 220 and 300, and the preset first threshold (effective completion unit quantile) is 200, and at this time, each distribution unit in the first distribution range needs to be judged according to the effective completion unit data of each distribution unit and the preset first threshold, so as to obtain a target sub-area unit, and the target sub-area unit is deleted.

The specific judgment basis is as follows: and judging whether the effective completion single data is smaller than a first threshold value, if so, taking the delivery unit corresponding to the effective completion single data as a target delivery unit, and deleting the target delivery unit. The effective completion unit quantile number reflects a standard quantity that all merchant orders corresponding to a certain distribution site can effectively complete distribution, the standard quantity is the lowest quantity which can enable a target object to be profitable, and when the effective completion unit quantity of a distribution unit is lower than the standard quantity, the distribution efficiency of the distribution unit is low, the input-output benefits of merchants or the distribution sites are not met, and therefore the effective completion unit quantile number can be eliminated.

According to the above judgment basis, it is judged that the delivery unit a1 is in accordance with the judgment basis and the delivery unit a2 and the delivery unit A3 are not in accordance with the judgment basis, so when determining the delivery range, it is necessary to delete the delivery unit a1 in the first delivery range, and keep the delivery unit a2 and the delivery unit A3, and if the remaining delivery units are not in accordance with the judgment basis, in the first delivery range, the delivery unit a1 is removed, and the area formed by the remaining delivery units, the delivery unit a2 and the delivery unit A3 is the second delivery range of the a merchant.

As can be seen, compared with the prior art, in the present embodiment, by deleting the area with low distribution efficiency (i.e., the target distribution unit) in the original distribution range, a more reasonable distribution range is determined, and the distribution quality is improved.

A third embodiment of the present invention provides a method for generating a delivery range, as shown in fig. 3-1, including:

step 301, obtaining order characteristic data of each delivery unit in a first delivery range of a target object;

step 302, judging whether the ratio of the overtime unit quantity to the effective completion unit quantity of the distribution unit is larger than an overtime ratio proportion reference;

step 303, if the ratio of the overtime unit amount to the effective completion unit amount of the distribution unit is greater than the overtime ratio proportion reference, determining the distribution unit as a target distribution unit;

in step 304, the target delivery unit is deleted from the first delivery range, and a second delivery range of the target object is generated.

In this embodiment, the first threshold includes a timeout single proportion criterion, the order characteristic data includes a timeout single quantity and an effective completion single quantity, the timeout single quantity refers to the number of orders that are not sent to the user within a specified time and result in timeout, as shown in fig. 3-2, the timeout single proportion criterion is obtained by the following method:

step 3021, counting the effective finished orders and the overtime orders of all the objects within a historical time period of the delivery site corresponding to the target object;

step 3022, taking the ratio of the overtime order quantity to the effective finished order quantity of all orders as the overtime order proportion reference.

Specifically, also taking the takeout industry as an example, when determining the overtime single proportion reference, firstly, counting the effective finished single quantity and the overtime single quantity in the takeout order data of all the merchants in a historical time period of the delivery site corresponding to the merchant, calculating the ratio between the overtime single quantity and the effective finished single quantity, and taking the ratio as the overtime single proportion reference. It should be noted that whether the order delivery is overtime or not may be determined according to actual delivery requirements, for example, for an order which is requested by a customer to be delivered within a time limit, whether the order is a time-out order or not may be determined according to whether the delivery time exceeds the requested time limit, and for some orders which are delivered in the same city or have no exact time limit, the time limit requirements specified by the operator may be determined, for example, the current day delivery or the next day delivery, all the time-out orders should be valid completion orders, and the order which is not completed finally does not fall within the time-out order range.

For example, if the number of valid finished orders of the takeout orders of all the merchants in the past 30 days by the delivery site corresponding to the merchant B is 1 ten thousand, and the number of overtime orders is 0.15 ten thousand, the overtime order proportion criterion determined according to the present embodiment is 0.15. If the existing distribution range of the B merchant is known, the existing distribution range is used as a first distribution range, and the first distribution range is divided into a plurality of grids with the same size, each grid is a distribution unit, and assuming that the first distribution range is composed of a distribution unit B1, a distribution unit B2, a distribution unit B3, and a distribution unit B4, the order distribution data of the distribution units are processed to obtain the timeout single rate (i.e., the ratio between the timeout single quantity and the effective completion single quantity) of each distribution unit, for example, the timeout single rate of the distribution unit B1 is 0.1, the timeout single rate of the distribution unit B2 is 0.2, the timeout single rate of the distribution unit B3 is 0.12, and the timeout single rate of the distribution unit B4 is 0.08. Determining the delivery unit as the target delivery unit if the ratio of the overtime unit amount to the effective completion unit amount of the delivery unit is greater than the overtime ratio reference according to the step 303, wherein the overtime unit rate of the delivery unit B2 is 0.2 and is greater than the overtime ratio reference 0.15, and the delivery unit B2 is deleted in the first delivery range according to the judgment principle; the overtime unit rates of the distribution units B1, B3 and B4 are all smaller than the overtime rate reference 0.15, and if the judgment principle is not met, the three distribution units are kept in the distribution range.

Compared with the prior art, the method and the device have the advantages that the regions with poor distribution quality (namely the target sub-region units) in the original distribution range are deleted, the more reasonable distribution range is determined, and the distribution efficiency is improved.

A fourth embodiment of the present invention provides a method for generating a delivery range, as shown in fig. 4-1, including:

step 401, obtaining order characteristic data of each delivery unit in a first delivery range of a target object;

step 402, judging whether the ratio of the overtime unit quantity to the effective completion unit quantity in the preset duration of the distribution unit is greater than an overtime ratio reference or not;

step 403, if the ratio of the overtime unit amount to the effective completion unit amount within the preset duration of the distribution unit is greater than the overtime ratio proportion reference, determining the distribution unit as a target distribution unit;

in step 404, the target delivery unit is deleted from the first delivery range, and a second delivery range of the target object is generated.

In this embodiment, the first threshold includes an overtime single proportion standard within a preset duration, the order characteristic data includes an overtime single quantity and an effective completion single quantity within the preset duration, the overtime single quantity within the preset duration refers to the number of orders that are not sent to the user within a preset time and result in overtime, and as shown in fig. 4-2, the overtime single proportion standard within the preset duration is obtained by the following method:

step 4021, counting effective completion orders of all objects in a historical time period and overtime orders in a preset time period of a distribution site corresponding to a target object;

step 4022, taking the ratio of the overtime single amount in the preset duration of all orders to the effective finished single amount as the standard of the overtime single proportion in the preset duration.

In practice, in the field of order distribution, especially in the distribution of take-out orders, for orders that are required by a time limit (orders that are delivered on a non-current day or delivered on a next day), a preset time period is used as a minimum criterion for judging whether the orders are overtime, and 60 minutes can be preset generally, that is, a take-out order is not delivered within 60 minutes and is marked as an order with an overtime. When the overtime order proportion reference in the preset time is determined, firstly, the effective completion order quantity and the overtime order quantity in the preset time are counted in the takeout order data of all objects of the distribution site corresponding to the commercial tenant in a historical time period, the ratio of the overtime order quantity and the effective completion order quantity in the preset time period is calculated, and the ratio is used as the overtime order proportion reference in the preset time period.

For example, if the number of valid completion orders of the takeout orders of all the merchants C in the last 30 days is 1 ten thousand, and the number of the over 60-minute orders is 0.25 ten thousand, the over 60-minute order proportion criterion determined according to the present embodiment is 0.25.

If the existing distribution range of the C merchant is known, the existing distribution range is used as a first distribution range, the first distribution range is divided into a plurality of grids with the same size, each grid is a distribution unit, the first distribution range is assumed to be composed of 87 distribution units, order distribution data of the 87 distribution units are processed to obtain effective completion single quantity and over 60-minute single quantity of each distribution unit, and the ratio of the effective completion single quantity to the over 60-minute single quantity of each distribution unit is calculated, for example, the ratio of the distribution unit C1 is 0.1, the ratio of the distribution unit C2 is 0.3, the ratio of the distribution unit C3 is 0.2, and the ratio of the distribution unit C4 is 0.35. According to the fact that the ratio of the overtime unit amount to the effective completion unit amount in the preset duration of the distribution unit is greater than the overtime ratio reference, the distribution unit is determined as the target distribution unit, and the ratio of the effective completion unit amount to the unit amount exceeding 60 minutes of the distribution unit C2 is 0.3 and is greater than the overtime ratio reference 0.25, which is described in the step 203; the ratio of the effective finished single quantity to the over 60 minute single quantity of the area unit C4 is 0.35, which is greater than the overtime rate reference 0.25, and the distribution units C2 and C4 are deleted in the first distribution range according to the judgment principle; the ratios of the effective completion single quantity and the over 60-minute single quantity of the distribution units C1 and C3 are both less than the over 60-minute single proportion standard of 0.25, and if the judgment principle is not met, the two distribution units are kept in the distribution range.

Compared with the prior art, the overtime single occupation ratio standard in the preset time length is a subset of the overtime single occupation ratio standard, and the order distribution efficiency of the order of a certain distribution unit can be reflected more accurately, so that the time length range is narrowed, a more accurate distribution range is obtained, and the distribution efficiency is improved.

It should be noted that, in the second to fourth embodiments, if a certain delivery unit meets at least one of the determination principles of the three embodiments, the delivery unit should be deleted in the first delivery range; if a certain delivery unit does not meet any of the criteria in the three embodiments, the delivery unit is one of the delivery units in the generated new delivery range. For example, in the third embodiment, the first distribution range is B5 … … B87 except for the distribution unit B1, the distribution unit B2, and the distribution unit B3 and the distribution unit B4, and the other distribution units do not meet any judgment rule, so that the second distribution range of the B business is a range formed by the distribution units B4 … … B87.

In addition, the above embodiments are merely representative preferred embodiments, and it should be understood by those skilled in the art that the present invention is not limited to the judgment principles illustrated in the above embodiments, and may also be judged by other factors having influences on the delivery quality or the delivery efficiency, and will not be described herein again.

A fifth embodiment of the present invention provides a method for generating a distribution range, as shown in fig. 5-1, including:

step 501, obtaining order characteristic data of each delivery unit in a first delivery range of a target object;

step 502, obtaining target delivery units from each delivery unit according to order characteristic data and a preset first threshold;

step 503, deleting the target delivery unit in the first delivery range, and generating a second delivery range of the target object;

step 504, acquiring a target area corresponding to the target object, and determining a plurality of sub-area units in the target area, wherein the target area is larger than a first distribution range, and the plurality of sub-area units are located in the target area outside the first distribution range;

step 505, obtaining an expanded sub-region unit according to the attribute information and the position information of the target object, the position information of the sub-region unit and a preset distribution prediction model;

step 506, add the expanded sub-region unit to the second delivery range of the target object.

In this embodiment, the target area of the target object is larger than the first delivery range of the target object, the set of area ranges to which the general order of the merchant is delivered is usually used as the first delivery range, the target area of the merchant is determined by the set of area ranges to which the special order of the merchant (such as some in-town delivery orders, running leg orders, etc.) is required to be delivered, and the determined target area should be larger than the first delivery range. For example, the distribution areas of the general orders which are usually taken by the target object are all within a distance range of 3 kilometers, a regular or irregular shaped geographical range within the distance range of 3 kilometers is taken as the first distribution range of the target object, the distribution areas of the special orders which are taken by the target object outside the general orders may be within a longer distance range, for example, within a distance range of 5 kilometers, and a regular or irregular shaped geographical range within a distance range of 5 kilometers is taken as the target area of the target object. The target area is divided into a plurality of grids with the same or different sizes according to the method of the above embodiment, and the grids in the target area outside the first distribution range are called sub-area units.

Further, in step 505, "obtain an expanded sub-region unit according to the attribute information and the position information of the target object, the position information of the sub-region unit, and the preset distribution prediction model", as shown in fig. 5-2, the method specifically includes:

step 5051, determining a predicted order quantity of a certain sub-region unit according to the attribute information of the target object and a preset first delivery prediction model;

step 5052, when the predicted order quantity is larger than the preset order quantity reference, determining the predicted delivery duration of the sub-area unit according to the position information of the target object, the position information of the sub-area unit and a preset second delivery prediction model;

in step 5053, when the predicted delivery duration is less than the average delivery duration, the sub-region unit is treated as an expanded sub-region unit.

Specifically, in determining which sub-area units can be added as expanded sub-area units in the second delivery range of the target object, to compensate for the reduced order size of the previously deleted delivery unit, an order size reference (e.g., order size of 0) and an average delivery duration (e.g., 45 minutes) are set, a predicted order size for the sub-area unit is first determined, and when the predicted order size for the sub-area unit is greater than 0, i.e., the sub-area unit is predicted to be ordered, then the predicted delivery duration for the sub-area unit is determined, when the predicted delivery duration of the sub-area unit is also less than 45 minutes, the system considers that the sub-area unit will not only generate an order, and the distribution efficiency of the order is relatively high, the sub-area unit is used as an expanded sub-area unit to be added into a second distribution range of the target object so as to generate an intelligent distribution range of the target object. It should be noted that the present embodiment only provides a way to determine the expansion sub-area unit, and the order quantity reference and the average delivery duration may also be set to other values according to the actual demand and the delivery pressure, and those skilled in the art should understand that other ways to determine the expansion sub-area unit may also be adopted to meet the needs of different delivery scenarios, and no additional description is made here.

After the areas with low delivery efficiency and poor delivery quality (i.e. the target sub-area units) are removed, although the accuracy of the delivery range is improved, the number of orders is reduced, but the present embodiment aims to compensate for the reduction of the number of orders, and expands the areas with orders and not with low delivery efficiency and poor delivery quality to compensate for the reduction of the number of orders of the target object caused by deleting the delivery units.

In addition, in practical application, for a sub-area unit with a large order quantity, no matter how bad the optimal distribution rate is, the sub-area unit is still kept not to be removed, because a user can keep a high order under a low optimal distribution rate, which indicates that the user can endure low distribution quality, and the distribution quality of the sub-area unit can be improved by other means.

A sixth embodiment of the present invention provides a method for generating a delivery range, as shown in fig. 6, including:

601, obtaining order characteristic data of each delivery unit in a first delivery range of a target object;

step 602, obtaining target delivery units from each delivery unit according to order characteristic data and a preset first threshold;

step 603, deleting the target delivery unit in the first delivery range, and generating a second delivery range of the target object;

step 604, acquiring a target area corresponding to the target object, and determining a plurality of sub-area units in the target area, wherein the target area is larger than a first distribution range, and the plurality of sub-area units are located in the target area outside the first distribution range;

step 607, inputting the attribute information of the target object in the trained first distribution prediction model to obtain the predicted order quantity of the sub-region unit;

step 610, when the predicted order quantity is larger than the preset order quantity reference, inputting the position information of the target object and the position information of a certain sub-area unit in the trained second distribution prediction model to obtain the predicted distribution duration of the sub-area unit;

step 611, when the predicted delivery duration is less than the average delivery duration, taking the sub-region unit as an expanded sub-region unit;

step 612, add the expanded sub-region unit to the second delivery range of the target object.

It should be noted that, preferably, before step 605, the method of this embodiment further includes:

step 605, constructing a first delivery prediction model, wherein the first delivery prediction model comprises a first prediction function, and the first prediction function takes the attribute information of the object and the order characteristic data of the sub-area unit as parameters;

step 606, according to a machine learning algorithm, training a first delivery prediction model by using all order data of a city where a target object is located in a historical time period;

preferably, before step 610, the method further comprises:

step 608, constructing a second distribution prediction model, where the second distribution prediction model includes a second prediction function, and the second prediction function takes the position information of the object and the position information of the sub-area unit as parameters;

step 609, training a second delivery prediction model according to a machine learning algorithm by using all order data of a city where the target object is located in a historical event section;

the embodiment provides a specific method for determining a predicted order quantity and a predicted delivery time length of a certain sub-region, and specifically, when determining the predicted order quantity of the certain sub-region, a first delivery prediction model is first constructed, the first delivery prediction model includes a first prediction function, the first prediction function is constructed by using attribute information of all objects in a business circle as parameters, for example, information such as a class, a price (a unit price of a customer), a brand (whether the brand is starbucks or not) of the business and the like as parameters, and then the first delivery prediction model is constructed, further, different weight values can be respectively given to the parameters to embody the influence of the parameters when predicting whether an order is generated by a certain sub-region unit; subsequently, the first delivery prediction model may be trained by using order characteristic data of all merchant orders in a historical time period (for example, the past 60 days) in city-wide, and the first delivery prediction model is trained by using city-wide data because the delivery range needs to be expanded in this embodiment, the sub-area unit required for expansion may be located outside the business circle where the target object is located, and if training is performed only by using order data of the business circle where the target object is located, the expansion sub-unit outside the business circle may be omitted. During training, for the distribution units within the first distribution range, the GBDT mode may be adopted for training and predicting, and for the (sub-region units) outside the first distribution range and within the target region, the Linear Regression algorithm may be adopted for training and predicting, where the GBDT and Linear Regression algorithm all belong to a relatively mature machine learning algorithm, which is not described herein;

after training is completed, we can obtain the attribute label of each delivery unit and sub-area unit, for example, the attribute label given to the delivery unit X according to the data such as effective completion list is brand coffee, and the guest unit price is 30-40 yuan; after the first distribution prediction model is trained, the attribute information of the target object, such as the order commodity class, the order unit price, the merchant brand and other information of the merchant are input into the trained first distribution prediction model, and after the model is used for finding out the corresponding attribute label, the sub-area unit can be predicted not to generate an order or how many orders can be generated by the sub-area unit.

Subsequently, when the predicted order quantity is larger than the preset order quantity reference, the predicted delivery time length of the sub-area unit needs to be further determined, so as to select the sub-area unit with shorter delivery time length (higher delivery efficiency) to join in the second delivery range. Specifically, a second delivery prediction model is constructed, the second delivery prediction model includes a second prediction function, the second prediction function takes the position information of the object and the position information of the sub-area unit as parameters, such as the longitude and latitude of a merchant and the longitude and latitude of a user, in practice, factors such as the speed of a rider and the weather condition may also be considered, the second prediction function is constructed by taking the parameters as parameters, and then the second delivery prediction model is constructed, further, different weight values can be respectively given to the parameters, so as to embody the influence of the parameters in predicting the order delivery duration of a certain sub-area unit; subsequently, the first distribution prediction model may be trained by using distribution duration data of orders of all merchants in a historical time period (for example, 60 days in the past) in a city, where the distribution duration data includes information such as a merchant ID of an order, a sub-area unit ID where the order is located, and an order distribution duration, and the first distribution prediction model is trained by using the city-wide data, because the distribution range needs to be expanded in this embodiment, the sub-area unit needed for expansion may be located outside a business circle where the target object is located, and if training is performed only with order data of the business circle where the target object is located, the expansion sub-unit outside the business circle may be omitted. The training can be performed by using an xgboost algorithm, which also belongs to a relatively mature machine learning algorithm and is not described herein.

After the second distribution prediction model is trained, the position information of the target object and the position information of the sub-area unit are input into the trained second distribution prediction model, so that the distribution time length of the sub-area unit can be predicted, and whether the sub-area unit is added into the second distribution range is determined by judging whether the predicted distribution time length of the sub-area unit exceeds the preset average distribution time length.

The embodiment provides a method for calculating the predicted order quantity and the predicted delivery time length of a certain sub-area unit, and the most optimized predicted value can be obtained through the method, so that the optimal order delivery range can be generated. It will be appreciated by those skilled in the art that other methods may be used to perform the calculation, and the examples listed in the present embodiment are not intended to be limiting.

A seventh embodiment of the present invention provides a method for generating a delivery range, as shown in fig. 7-1, including:

step 701, obtaining order characteristic data of each delivery unit in a first delivery range of a target object;

step 702, obtaining target delivery units from each delivery unit according to order characteristic data and a preset first threshold;

step 703 of deleting the target delivery unit within the first delivery range and generating a second delivery range of the target object;

step 704, obtaining a target area corresponding to the target object, and determining a plurality of sub-area units in the target area, wherein the target area is larger than a first distribution range, and the plurality of sub-area units are located in the target area outside the first distribution range;

step 705, obtaining an expanded sub-region unit according to the attribute information and the position information of the target object, the position information of the sub-region unit and a preset distribution prediction model;

step 706, adding the expanded sub-region unit into a second distribution range of the target object;

step 707, acquiring a geo-fence including a preset keyword;

step 708, judging each distribution unit and sub-area unit in the second distribution range according to the geo-fence to obtain an invalid unit;

in step 709, the invalid unit is deleted in the second delivery range.

In this embodiment, steps 701-706 are described in the above embodiments, and are not described herein. In step 707, the keywords may be rivers, seas, scenic spots, parks, cemeteries, etc., where no orders are present, and therefore called as invalid areas, and it is determined whether the distribution units in the first distribution range and each sub-area unit in the target area include rivers, seas, parks, cemeteries, or whether the distribution units in the first distribution range and each sub-area unit in the target area are located entirely within rivers, seas, parks, cemeteries, and if so, the invalid sub-area units including the invalid areas are removed from the second distribution range.

As shown in fig. 7-2, the shape of the water drop at the center is the position of the target object (merchant), all grids at the periphery of the target object are sub-area units, where the set of gray grid, dark gray grid, and white grid is the first distribution range, the dark gray grid (for example, the grid with identification number 2) in the first distribution range is the target distribution unit, that is, the unit that needs to be deleted in step 703, the white grid (for example, the grid with identification number 3) in the first distribution range is the invalid sub-area unit, that is, the unit that needs to be removed in step 709, and the other gray grids (for example, the grid with identification number 1) in the first distribution range are distribution units that need to be retained in the distribution range; the set of light gray and white grids outside the first delivery range is the target area, the light gray grid (for example, the grid with the identification number 4) in the target area is the expanded sub-area unit, i.e., the unit to be added in step 706, the white grid (for example, the grid with the identification number 5) in the target area is also the invalid sub-area unit, i.e., the unit to be removed in step 709, and the dark shaded portion shown in fig. 7-3 is the final intelligent delivery range.

In the present embodiment, the new delivery range is generated by the delivery unit with low delivery efficiency and poor delivery quality by eliminating the first delivery range from the existing delivery range; finding out sub-area units which generate orders and have delivery duration meeting the expansion strategy in the target area so as to make up for the reduced number of the sub-area units; and eliminating invalid units of the first distribution range and the target area, wherein the final filtered merchant unit collection is the generated new distribution range.

An eighth embodiment of the present invention provides a distribution range generation device, as shown in fig. 8, including:

a data obtaining module 801, configured to obtain order feature data of each delivery unit in a first delivery range of a target object;

the grid judgment module 802 obtains a target delivery unit from each delivery unit according to the order characteristic data and a preset first threshold;

the grid filtering module 803 deletes the target delivery unit within the first delivery scope.

Further, the data obtaining module 801 obtains order feature data of each sub-area unit in the first area of the target object, including extracting order data of the first distribution range from the database; and screening the order data to obtain order characteristic data of each distribution unit. The order characteristic data can provide characteristic data with the most influence, and unreasonable distribution units can be eliminated conveniently through calculation.

As a further improvement, the first threshold of the grid determining module 802 includes an effective completion unit quantile, the order characteristic data includes an effective completion unit quantity, and the delivery unit is determined according to the order characteristic data and a preset first threshold to obtain a target delivery unit, specifically: judging whether the effective completion unit quantity of the distribution unit is smaller than the effective completion unit quantile or not; and if the effective completion unit quantity of the delivery unit is less than the effective completion unit quantile, determining the delivery unit as the target delivery unit. When the effective completion single amount of a certain delivery unit is less, the delivery efficiency of the sub-area unit is considered to be lower, and the delivery range of the target object can be optimized by removing the delivery unit from the delivery range.

As a further improvement, the first threshold of the grid determining module 802 includes an overtime list proportion standard, the order characteristic data includes an overtime list quantity and an effective finished list quantity, and the delivery unit is determined according to the order characteristic data and a preset first threshold to obtain a target delivery unit, specifically: judging whether the ratio of the overtime single quantity to the effective completion single quantity of the distribution unit is larger than an overtime ratio reference or not; and if the ratio of the distribution units is larger than the overtime ratio reference, determining the distribution unit as a target distribution unit. When the overtime probability of a certain distribution unit is higher, the distribution quality of the distribution unit is considered to be poor, and the rationality of the distribution range of the merchant can be improved by removing the sub-area unit.

In addition, the timeout duty cycle reference is obtained according to the following method: counting effective finished orders and overtime orders of all orders in a historical time period in a city where the target object is located; and taking the ratio of the overtime order quantity of all orders to the effective finished order quantity as the overtime order proportion reference.

As a further improvement, the first threshold of the grid determining module 802 includes a standard exceeding a preset time length and an order form ratio, the order form feature data includes a single quantity exceeding the preset time length and a single quantity effective, and the delivering unit is determined according to the order form feature data and the preset first threshold to obtain a target delivering unit, specifically: judging whether the ratio of the single quantity exceeding the preset time length of the distribution unit to the single quantity effectively completed is larger than the reference of the ratio exceeding the preset time length of the single quantity; and if the ratio of the distribution units is larger than the preset time-length single-occupation ratio reference, determining the distribution units as target distribution units. When the overtime probability in the preset duration of a certain distribution unit is larger, the distribution quality of the sub-area unit is considered to be poor, and the rationality of the distribution range of the merchant can be further improved by removing the sub-area unit.

In addition, the single proportion reference exceeding the preset time length is obtained according to the following method: counting effective completion orders and units exceeding preset time duration of all orders in a city where the target object is located within a historical time period; and taking the ratio of the units with the exceeding preset time length of all orders to the effective completion units as the reference of the ratio of the units with the exceeding preset time length.

As a further improvement, in order to compensate for the reduced number of delivery units, the delivery range generation device of the present embodiment further includes:

the area obtaining module 804 is configured to obtain a target area corresponding to the target object, and determine a plurality of sub-area units in the target area, where the target area is larger than the first distribution range, and the plurality of sub-area units are located in the target area outside the first distribution range;

the expansion judgment module 805 obtains an expanded sub-region unit according to the attribute information and the position information of the target object, the position information of the sub-region unit, and a preset distribution prediction model;

the grid expansion module 806 adds the expanded sub-region units to the second delivery range of the target object. After unreasonable areas with low delivery efficiency and poor delivery quality are removed, in order to make up for the reduced order quantity, the areas are expanded, the delivery range is further expanded, and the more optimal delivery range is finally determined.

In addition, the expansion determining module 805 obtains an expanded sub-region unit according to the attribute information and the position information of the target object, the position information of the sub-region unit, and the preset delivery prediction model, and specifically includes: determining a predicted order quantity of a certain sub-area unit according to the attribute information of the target object and a preset first delivery prediction model; when the predicted order quantity is larger than the preset order quantity reference, determining the predicted delivery duration of the sub-area unit according to the position information of the target object, the position information of the sub-area unit and a preset second delivery prediction model; and when the predicted distribution time length is less than the average distribution time length, taking the sub-area unit as an expanded sub-area unit. Firstly, whether a certain sub-area unit can generate an order is judged, when the order can be generated is predicted, the distribution time length of the sub-area unit is judged, if the distribution time length is shorter, the distribution quality and the distribution efficiency of the sub-area unit are higher, and the sub-area unit can be used as an expanded area to be added into the distribution range of a target object.

In addition, the expansion determining module 805 determines the predicted order amount of a certain sub-area unit according to the attribute information of the target object and a preset first delivery prediction model, and specifically includes: constructing a first delivery prediction model, wherein the first delivery prediction model comprises a first prediction function, and the first prediction function takes attribute information of an object as a parameter; according to a machine learning algorithm, training a first delivery prediction model by using order characteristic data of all orders of a city where a target object is located in a historical time period; and inputting the attribute information of the target object in the trained first delivery prediction model to obtain the predicted order quantity of the sub-region unit. The most accurate predicted quantity is obtained by intelligently training the order quantity, and when the predicted order quantity of a certain sub-area unit is large, the sub-area unit can be added into a distribution range.

In addition, the expansion determining module 805 determines the predicted delivery duration of the sub-area unit according to the position information of the target object, the position information of the sub-area unit, and a preset second delivery prediction model, and specifically includes: constructing a second delivery prediction model, wherein the second delivery prediction model comprises a second prediction function, and the second prediction function takes the position information of the object and the position information of the sub-area unit as parameters; according to a machine learning algorithm, training a second delivery prediction model by using delivery duration data of all orders of a city where the target object is located in a historical event section; and inputting the position information of the target object and the position information of a certain sub-area unit in the trained second distribution prediction model to obtain the predicted distribution time length of the sub-area unit. And obtaining the most accurate predicted time length according to the intelligent training, and adding a sub-region unit into a distribution range when the predicted time length of order distribution of the sub-region unit is smaller.

As a further improvement, the delivery range generation device according to the present embodiment further includes, when generating the delivery range:

a fence acquisition module 807 that acquires a geo-fence including a preset keyword;

an invalid judgment module 808, configured to judge each distribution unit and sub-area unit in the second distribution range according to the geo-fence, to obtain an invalid unit;

the invalidation deleting module 809 deletes the invalidation unit in the second distribution range. By deleting invalid areas where logistics such as rivers, lakes and seas, scenic spots, parks, cemeteries and the like cannot be distributed, the order distribution range is further optimized.

The ninth embodiment of the present invention relates to an electronic device, and the electronic device of the present embodiment may be a terminal device, such as a mobile phone, a tablet computer, and the like, and may also be a server on a network side.

As shown in fig. 9, the electronic device: at least one processor 901; and, memory 902 communicatively connected to at least one processor 901; and a communication component 903 communicatively coupled to the scanning device, the communication component 903 receiving and transmitting data under the control of the processor 901; wherein the memory 902 stores instructions executable by the at least one processor 901, the instructions being executable by the at least one processor 901 to implement:

the method comprises the steps of obtaining order characteristic data of each delivery unit in a first delivery range of a target object; obtaining target delivery units from each delivery unit according to the order characteristic data and a preset first threshold value; deleting the target delivery unit in the first delivery range;

acquiring a target area corresponding to a target object, and determining a plurality of sub-area units in the target area, wherein the target area is larger than a first distribution range, and the plurality of sub-area units are positioned in the target area outside the first distribution range; obtaining an expanded sub-region unit according to the attribute information and the position information of the target object, the position information of the sub-region unit and a preset distribution prediction model; adding the expanded sub-area unit into a second distribution range of the target object, wherein the second distribution range is the distribution range generated by deleting the target distribution unit in the first distribution range;

acquiring a geo-fence comprising preset keywords; judging each distribution unit and sub-area unit in the second distribution range according to the geo-fence to obtain an invalid unit; the invalid unit is deleted in the second delivery range. By deleting invalid areas where logistics such as rivers, lakes and seas, scenic spots, parks, cemeteries and the like cannot be distributed, the order distribution range is further optimized.

Specifically, one processor 901 is taken as an example in fig. 9. The processor 901 and the memory 902 may be connected by a bus or by other means, and fig. 9 illustrates the connection by the bus as an example. Memory 902, which is a non-volatile computer-readable storage medium, may be used to store non-volatile software programs, non-volatile computer-executable programs, and modules. The processor 901 executes various functional applications and data processing of the device, that is, implements the above-described distribution range generation method, by executing nonvolatile software programs, instructions, and modules stored in the memory 902.

The memory 902 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required for at least one function; the storage data area may store a list of options, etc. Further, the memory 902 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some embodiments, the memory 902 may optionally include memory located remotely from the processor 901, which may be connected to an external device via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

One or more modules are stored in the memory 902 and when executed by the one or more processors 901 perform the method of generating a delivery range in any of the method embodiments described above.

The product can execute the method provided by the embodiment of the application, has corresponding functional modules and beneficial effects of the execution method, and can refer to the method provided by the embodiment of the application without detailed technical details in the embodiment.

In the embodiment, the regions with low distribution efficiency and poor distribution quality are deleted according to the filtering strategy, new sub-region units are added according to the expansion strategy, and invalid regions are removed to determine a more reasonable distribution range, so that the problems that some orders are difficult to distribute or the regions with poor distribution quality cannot be randomly changed due to static time circle, the formed distribution range is not optimized enough, and the order distribution efficiency is influenced in the existing distribution range determining method are solved.

A tenth embodiment of the invention is directed to a non-volatile storage medium storing a computer-readable program for causing a computer to perform some or all of the above method embodiments.

That is, as can be understood by those skilled in the art, all or part of the steps in the method for implementing the embodiments described above may be implemented by a program instructing related hardware, where the program is stored in a storage medium and includes several instructions to enable a device (which may be a single chip, a chip, or the like) or a processor (processor) to execute all or part of the steps of the method described in the embodiments of the present application. And the aforementioned storage medium includes: a U disk, a removable hard disk, a Read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

It will be understood by those of ordinary skill in the art that the foregoing embodiments are specific examples for carrying out the invention, and that various changes in form and details may be made therein without departing from the spirit and scope of the invention in practice.

The embodiment of the application discloses A1. a method for generating a distribution range, which comprises the following steps:

the method comprises the steps of obtaining order characteristic data of each delivery unit in a first delivery range of a target object;

obtaining a target delivery unit from the delivery units according to the order characteristic data and a preset first threshold value;

and deleting the target delivery unit in the first delivery range.

A2. The method for generating a distribution range according to a1, where the first threshold includes an effective completion unit quantile, the order feature data includes an effective completion unit quantity, and the obtaining a target distribution unit from the distribution units according to the order feature data and a preset first threshold specifically includes:

determining whether the effective completion unit count of the delivery unit is less than the effective completion unit quantile;

and if the effective completion unit quantity of the distribution unit is smaller than the effective completion unit quantile, determining the distribution unit as the target distribution unit.

A3. According to the distribution range generating method described in a2, the effective completion unit quantile is obtained by:

determining a distribution range set of all objects in a historical time period by a distribution site corresponding to the target object;

sorting the distribution units in the distribution range set according to the size of the effective completion single quantity;

and determining a corresponding reference distribution unit in the sequence according to a preset percentage reference, and taking the effective completion unit quantity of the reference distribution unit as the effective completion unit quantile.

A4. According to the method for generating a distribution range described in a1, the first threshold includes a timeout order proportion reference, the order feature data includes a timeout order quantity and an effective completion order quantity, and a target distribution unit is obtained from the distribution units according to the order feature data and a preset first threshold, specifically:

judging whether the ratio of the overtime single quantity to the effective completion single quantity of the distribution unit is larger than the overtime ratio proportion reference or not;

and if the ratio of the delivery units is larger than the timeout rate ratio reference, determining the delivery units as the target delivery units.

A5. According to the method for generating a distribution range described in a4, the timeout single proportion criterion is obtained according to the following method:

counting effective finished orders and overtime orders of all objects in a historical time period of a delivery site corresponding to the target object;

and taking the ratio of the overtime single amount of all orders to the effective finished single amount as the standard of the overtime single proportion.

A6. According to the method for generating the distribution range described in a1, the first threshold includes a criterion exceeding a preset time-length sheet ratio, the order feature data includes a sheet amount exceeding a preset time length and an effective sheet amount, and a target distribution unit is obtained from the distribution units according to the order feature data and the preset first threshold, specifically:

judging whether the ratio of the single quantity of the super-preset duration to the single quantity of the effective completion of the distribution unit is greater than the single-proportion reference of the super-preset duration;

and if the ratio of the distribution units is greater than the preset duration single-proportion reference, determining the distribution units as the target distribution units.

A7. According to the method for generating the distribution range a6, the beyond-preset-duration single-proportion reference is obtained according to the following method:

counting the effective completion unit quantity and the unit quantity exceeding the preset time length of the orders of all the objects in a historical time period of the delivery station corresponding to the target object;

and taking the ratio of the units with the exceeding preset time length of all the orders to the effective completion units as the reference of the ratio of the units with the exceeding preset time length.

A8. The distribution range generation method according to any one of a1-a7, further comprising:

acquiring a target area corresponding to the target object, and determining a plurality of sub-area units in the target area, wherein the target area is larger than the first distribution range, and the plurality of sub-area units are located in the target area outside the first distribution range;

obtaining an expanded sub-region unit according to the attribute information and the position information of the target object, the position information of the sub-region unit and a preset distribution prediction model;

and adding the expanded sub-area unit into a second distribution range of the target object, wherein the second distribution range is the distribution range generated by deleting the target distribution unit in the first distribution range.

A9. According to the method for generating a distribution range described in A8, obtaining an expanded sub-region unit according to the attribute information and the position information of the target object, the position information of the sub-region unit, and a preset distribution prediction model, specifically:

determining the predicted order quantity of the sub-area unit according to the attribute information of the target object and a preset first delivery prediction model;

when the predicted order quantity is larger than a preset order quantity reference, determining the predicted delivery duration of the sub-area unit according to the position information of the target object, the position information of the sub-area unit and a preset second delivery prediction model;

and when the predicted distribution time length is less than the average distribution time length, taking the sub-area unit as the expanded sub-area unit.

A10. The method for generating a delivery range according to a9, the method further comprising:

constructing the first delivery prediction model, wherein the first delivery prediction model comprises a first prediction function, and the first prediction function takes attribute information of an object as a parameter;

according to a machine learning algorithm, training the first delivery prediction model by using order characteristic data of all orders in a historical time period of a city where the target object is located;

the determining the predicted order quantity of the sub-region unit according to the attribute information of the target object and a preset first delivery prediction model comprises:

and inputting the attribute information of the target object in the trained first distribution prediction model to obtain the predicted order quantity of the sub-region unit.

A11. The distribution range generation method according to A9A 10, the method further comprising:

constructing a second delivery prediction model, wherein the second delivery prediction model comprises a second prediction function, and the second prediction function takes the position information of the object and the position information of the sub-area unit as parameters;

according to a machine learning algorithm, training the second delivery prediction model by using delivery duration data of all orders in a historical time period of a city where the target object is located;

the determining the predicted delivery duration of the sub-area unit according to the position information of the target object, the position information of the sub-area unit and a preset second delivery prediction model includes:

and inputting the position information of the target object and the position information of a certain sub-area unit in the trained second distribution prediction model to obtain the predicted distribution time length of the sub-area unit.

A12. According to the method for generating a distribution range of any one of a1-a7, the obtaining of order characteristic data of each distribution unit in the first distribution range of the target object specifically includes:

extracting order data of the first distribution range from a database;

and screening the order data to obtain the order characteristic data of each distribution unit.

A13. The distribution range generation method according to A8, further comprising:

acquiring a geo-fence comprising preset keywords;

judging the distribution units and the sub-area units in the second distribution range according to the geo-fence to obtain invalid units;

deleting the invalid unit in the second delivery range.

The embodiment of the application discloses B14. a distribution range generation device, including:

the data acquisition module is used for acquiring order characteristic data of each distribution unit in a first distribution range of the target object;

the grid judgment module is used for obtaining a target delivery unit from the delivery units according to the order characteristic data and a preset first threshold value;

and the grid filtering module deletes the target delivery unit in the first delivery range.

The embodiment of the application discloses C15. an electronic device, comprising at least one processor;

a memory communicatively coupled to the at least one processor; and

the communication component is respectively in communication connection with the processor and the memory and receives and transmits data under the control of the processor; wherein the memory stores instructions executable by the at least one processor to implement:

and deleting the target delivery unit in the first delivery range.

C16. According to the electronic device described in C15, the first threshold includes an effective fraction completed number, the order feature data includes an effective fraction completed number, and the obtaining of the target delivery unit from the delivery unit according to the order feature data and the preset first threshold specifically includes:

C17. The electronic device of C16, wherein the effective completion monodispersion is obtained according to the following method:

C18. According to the electronic device described in C15, the first threshold includes a timeout sheet proportion criterion, the order feature data includes a timeout sheet quantity and an effective finished sheet quantity, and the obtaining of the target delivery unit from the delivery units according to the order feature data and the preset first threshold specifically includes:

C19. The electronic device of C18, wherein the timeout duty cycle reference is obtained according to the following method:

C20. According to the electronic device of C15, the first threshold includes a criterion exceeding a preset time-duration sheet-to-sheet ratio, the order feature data includes a unit amount exceeding a preset time duration and an effective unit amount, and the obtaining of the target delivery unit from the delivery unit according to the order feature data and the preset first threshold specifically includes:

C21. According to the electronic device of C20, the beyond preset duration single proportion reference is obtained according to the following method:

counting effective completion unit quantities and unit quantities exceeding a preset time length of orders of all objects in a historical time period of a distribution station corresponding to the target object;

C22. The electronic device of any of C15-C21, the processor further to perform:

C23. According to the electronic device of C22, obtaining an expanded sub-region unit according to the attribute information and the location information of the target object, the location information of the sub-region unit, and a preset delivery prediction model, specifically:

C24. The electronic device of C23, the processor further performing:

the step of determining a predicted order quantity of a certain sub-region unit by the processor executing the preset first distribution prediction model according to the attribute information of the target object and the sub-region unit specifically includes:

C25. The electronic device of C23, the processor further performing:

according to a machine learning algorithm, training the second delivery prediction model by using delivery duration data of all orders in a historical event section of a city where the target object is located;

the processor executes the predicted delivery duration of the sub-area unit according to the position information of the target object, the position information of the sub-area unit and a preset second delivery prediction model, and specifically includes:

C26. According to the electronic device of any one of C15-C21, the obtaining of the order feature data of each delivery unit within the first delivery range of the target object specifically includes:

extracting order data of the first distribution range from a database;

C27. The electronic device of C22, the processor further performing:

acquiring a geo-fence comprising preset keywords;

deleting the invalid unit in the second delivery range.

Disclosed in an embodiment of the present application is a nonvolatile storage medium for storing a computer-readable program for causing a computer to execute the distribution range generation method according to any one of a1 to a13.

Claims

1. A method for generating a delivery range, comprising:

dividing a first distribution range of a target object into at least two distribution units, and acquiring order characteristic data of each distribution unit in the first distribution range of the target object; wherein the order characteristic data at least comprises: an effective finished order amount and/or an overtime order amount for the order;

obtaining a target delivery unit from the delivery units according to the order characteristic data and a preset first threshold value; wherein the delivery efficiency and/or the delivery quality of the target delivery unit is lower than that of other delivery units;

deleting the target delivery unit in the first delivery range;

and after the target delivery unit is deleted, generating an expanded sub-area unit for the target object at least according to the attribute information and the position information of the target object, wherein the delivery efficiency of the expanded sub-area unit is greater than a set efficiency threshold and/or the delivery quality is greater than a set quality threshold.

2. The method for generating a distribution range according to claim 1, wherein the first threshold includes an effective completion unit quantile, the order feature data includes an effective completion unit quantity, and the obtaining of the target distribution unit from the distribution units according to the order feature data and the preset first threshold specifically includes:

3. The method of generating a delivery envelope of claim 2, wherein the number of effective completion monodispersions is obtained by:

4. The method for generating a delivery range according to claim 1, wherein the first threshold includes a timeout sheet proportion reference, the order characteristic data includes a timeout sheet quantity and an effective completion sheet quantity, and the target delivery unit is obtained from the delivery units according to the order characteristic data and a preset first threshold, specifically:

judging whether the ratio of the overtime single quantity to the effective completion single quantity of the distribution unit is larger than the overtime single proportion reference or not;

and if the ratio of the delivery units is larger than the overtime single-occupation-ratio reference, determining the delivery units as the target delivery units.

5. The method of generating a delivery area according to claim 4, wherein the timeout duty reference is obtained according to the following method:

and taking the ratio of the overtime single amount to the effective finished single amount of all orders as the standard of the overtime single proportion.

6. The method for generating a delivery range according to claim 1, wherein the first threshold includes a criterion exceeding a preset time-duration sheet-to-sheet ratio, the order characteristic data includes a sheet amount exceeding a preset time duration and an effective sheet amount, and the obtaining of the target delivery unit from the delivery units according to the order characteristic data and the preset first threshold specifically includes:

7. The delivery range generation method according to claim 6, wherein the beyond preset duration single proportion reference is obtained according to the following method:

and taking the ratio of the units with the exceeding preset time length of all orders to the effective completion units as the reference of the ratio of the units with the exceeding preset time length.

8. The distribution range generation method according to any one of claims 1 to 7, further comprising:

9. The method for generating a distribution range according to claim 8, wherein the expanded sub-region unit is obtained according to the attribute information and the position information of the target object, the position information of the sub-region unit, and a preset distribution prediction model, and specifically is:

10. The delivery range generation method according to claim 9, further comprising:

11. The delivery range generation method according to claim 9 or 10, further comprising:

12. The method for generating a distribution range according to any one of claims 1 to 7, wherein the obtaining of the order characteristic data of each distribution unit in the first distribution range of the target object specifically includes:

extracting order data of the first distribution range from a database;

13. The delivery range generation method according to claim 8, further comprising:

acquiring a geo-fence comprising preset keywords;

deleting the invalid unit in the second delivery range.

14. An apparatus for generating a delivery range, comprising:

the data acquisition module is used for dividing the first distribution range of the target object into at least two distribution units and acquiring order characteristic data of each distribution unit in the first distribution range of the target object; wherein the order characteristic data at least comprises: an effective finished order amount and/or an overtime order amount for the order;

the grid judgment module is used for obtaining a target delivery unit from the delivery units according to the order characteristic data and a preset first threshold value; wherein the delivery efficiency and/or the delivery quality of the target delivery unit is lower than that of other delivery units;

the grid filtering module deletes the target delivery unit in the first delivery range; and after the target delivery unit is deleted, generating an expanded sub-area unit for the target object at least according to the attribute information and the position information of the target object, wherein the delivery efficiency of the expanded sub-area unit is greater than a set efficiency threshold and/or the delivery quality is greater than a set quality threshold.

15. An electronic device comprising at least one processor;

a memory communicatively coupled to the at least one processor; and

deleting the target delivery unit in the first delivery range;

16. The electronic device according to claim 15, wherein the first threshold includes an effective finished single quantile, the order feature data includes an effective finished single quantity, and the target delivery unit is obtained from the delivery unit according to the order feature data and a preset first threshold, specifically:

17. The electronic device of claim 16, wherein the effective completion monodispersion is obtained according to the following method:

18. The electronic device according to claim 15, wherein the first threshold includes a timeout sheet proportion reference, the order feature data includes a timeout sheet quantity and an effective completion sheet quantity, and the obtaining of the target delivery unit from the delivery units according to the order feature data and a preset first threshold specifically includes:

19. The electronic device of claim 18, wherein the timeout duty cycle reference is obtained according to the following method:

20. The electronic device according to claim 15, wherein the first threshold includes a criterion exceeding a preset time duration and an order form ratio, the order feature data includes a form amount exceeding a preset time duration and an effective form amount, and the obtaining of the target delivery unit from the delivery units according to the order feature data and the preset first threshold specifically includes:

21. The electronic device of claim 20, wherein the beyond preset duration single proportion reference is obtained according to the following method:

22. The electronic device of any one of claims 15-21, wherein the processor further performs:

23. The electronic device according to claim 22, wherein the expanded sub-region unit is obtained according to the attribute information and the position information of the target object, the position information of the sub-region unit, and a preset delivery prediction model, and specifically is:

24. The electronic device of claim 23, wherein the processor further performs:

25. The electronic device of claim 23, wherein the processor further performs:

26. The electronic device according to any one of claims 15 to 21, wherein the obtaining of order characteristic data of each delivery unit within the first delivery range of the target object is specifically:

extracting order data of the first distribution range from a database;

27. The electronic device of claim 22, wherein the processor further performs:

acquiring a geo-fence comprising preset keywords;

deleting the invalid unit in the second delivery range.

28. A non-volatile storage medium storing a computer-readable program for causing a computer to execute the distribution range generation method according to any one of claims 1 to 13.