Order assignment method and system

By detecting the location of the dispatcher and calculating the order correlation degree, and optimizing the order assignment, the problems of unreasonable order assignment and shortage of dispatchers in the existing technology are solved, and the reasonable allocation of orders and the improvement of delivery staff satisfaction are achieved.

CN113723673BActive Publication Date: 2024-12-03ZHENGZHOU SHIKONG SUIDAO INFORMATION TECH CO LTD
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202110946875.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-08-18
Publication Date
2024-12-03
Estimated Expiration
2041-08-18

AI Technical Summary

Technical Problem

In the existing order delivery plan, the delivery staff may go to a location opposite to the starting point of the order, resulting in unreasonable order assignment, especially during holidays or shortage of delivery staff, resulting in delayed orders and failure to merge and deliver along the way, which wastes manpower and material resources.

Method used

By detecting the real-time and estimated location of the delivery staff, calculate the time and space correlation of the orders to be dispatched, bind similar orders, and query the list of optional delivery staff within the starting point setting range, optimize order assignments, and improve delivery staff Satisfaction and user satisfaction.

Benefits of technology

It realizes the rationality of order distribution, saves manpower and material resources, improves the satisfaction of delivery personnel and users, and reduces order delays, especially when there is a shortage of delivery personnel, promoting order grabbing through an incentive mechanism.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113723673B_ABST
    Figure CN113723673B_ABST
Patent Text Reader

Abstract

The present invention relates to an order assignment method and system, which detects the real-time position of a deliveryman in delivery, and estimates the trajectory of the deliveryman within a period of time according to the positions of the starting point and the end point and the delivery time; detects the real-time position of a deliveryman who is not in the delivery state; obtains a list of orders to be delivered within a period of time; calculates the time and space correlation of each order in the list of orders to be delivered, binds the orders with correlation exceeding a threshold into a group, and jointly assigns deliverymen; for each group of orders or a single unbound order, queries the deliverymen whose real-time positions are within the set range of the starting point and the deliverymen whose estimated positions are within the set range of the starting point at the start time, and forms a list of optional deliverymen; assigns deliverymen according to the deliverymen list until a deliveryman accepts the order. The delivery method of the present invention makes order assignment more reasonable; similar orders are bundled to save manpower and material resources.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of logistics delivery, and in particular to an order assignment method and system. Background Art

[0002] The current order delivery solution usually places orders based on real-time positioning results. The delivery person may go to a location opposite to the starting point of the order, which results in unreasonable order assignment.

[0003] Some orders are delivered in a rush mode, which results in a large number of people rushing to deliver some orders, while some orders are ignored. Especially during the peak order period during holidays or when there are few orders and delivery personnel, when an order is broadcast, there are no delivery personnel nearby to grab the order, causing a large number of order delays.

[0004] In addition, if orders that are on the way can be combined for delivery, it will definitely save manpower and material resources, and will also increase the satisfaction of the delivery staff.

[0005] In summary, how to distinguish different situations and optimize the dispatching strategy is a technical problem that needs to be solved urgently in this field. Summary of the invention

[0006] In view of the problems existing in the prior art, the present invention provides an order assignment method and system, which selects the delivery personnel in delivery based on the estimated location and selects the delivery personnel who are not in the delivery state based on the real-time location, so that the order assignment is more reasonable; similar orders are bundled to save manpower and material resources. At the same time, the satisfaction of users and delivery personnel is improved.

[0007] To achieve the above object, the present invention provides an order assignment method, comprising:

[0008] Detect the real-time location of the delivery person during delivery, and estimate the delivery person's trajectory over a period of time based on the locations of the starting and ending points and the delivery time;

[0009] Check the real-time location of the delivery personnel who are not in the delivery state;

[0010] Get a list of orders to be delivered within a certain period of time;

[0011] Calculate the time and space correlation of each order in the list of orders to be delivered, bind the orders with correlation exceeding the threshold into a group, and assign delivery personnel together;

[0012] For each group of orders or unbound single order, query the delivery personnel whose real-time location is within the set range of the starting point and the delivery personnel whose estimated location at the start time is within the set range of the starting point to form a list of optional delivery personnel;

[0013] Assign couriers according to the couriers list until a couriers accepts the order.

[0014] Furthermore, the trajectory of the deliveryman over a period of time is estimated, including: calculating the deliveryman's speed, calculating the deliveryman's position every k minutes according to the planned delivery path for the deliveryman and the deliveryman's speed, until the deliveryman reaches the target location, forming a trajectory over a period of time. After the deliveryman completes the delivery, his position is no longer estimated.

[0015] Furthermore, the calculation of the temporal and spatial correlation between each order and other orders includes: removing orders of types that are not allowed to be assigned at the same time, establishing feature vectors for clustering based on order attributes, the attributes include: starting location, target location, order item type, and start time, and for a cluster, the distance to other points is used as the correlation, and the orders corresponding to the points with a correlation less than a threshold are bundled. Furthermore, clustering is performed using K-MEANS, DBSCAN, or BRICH models.

[0016] Furthermore, the deliverymen in the optional deliverymen list are excluded from the deliverymen who do not accept orders, the deliverymen who have rejected orders in the last 6 hours, the deliverymen who have received the maximum number of orders, the deliverymen who have reached the maximum transportation capacity, the blacklisted deliverymen, and the deliverymen whose delivery types do not match. The remaining deliverymen are ranked according to the comprehensive index;

[0017] The calculation of the deliveryman's comprehensive index includes: Z = Q*k*(P+M+R+H)

[0018] P is user satisfaction, Q is the type of deliveryman dispatch, 1 for one order and -1 for multiple orders, M is the level score, k is the order type, 1 for one order and -1 for multiple orders, R is 0 for novice and 1 otherwise, H is 5 for the user's favorite runner. Furthermore, if the list of available deliverymen is empty, the setting range is expanded and the deliveryman is searched again. If the list of available deliverymen is still empty, the deliveryman is added to the assignment list;

[0019] After adding a reward amount to the orders in the assignment list, the order will be grabbed. If multiple people grab the order, the courier closest to the starting point will be selected; if no one grabs the order, the order will be dispatched manually.

[0020] Furthermore, after the order is delivered, the delivery person can choose to accept orders within a period of time, or accept part of a group of orders; for unaccepted orders, the next delivery person will be assigned according to the list of available delivery persons. If there is no next delivery person, the returned order will be set to the highest priority in the list of orders to be delivered and reassigned.

[0021] On the other hand, an order assignment system is provided, including a real-time location detection module, a location estimation module, a correlation calculation module, a delivery list generation module, and an assignment module;

[0022] The real-time location detection module locates all delivery personnel in real time and detects their real-time locations;

[0023] The location estimation module estimates the trajectory of the delivery person over a period of time based on the locations of the starting point and the end point, the real-time location, and the delivery time.

[0024] The correlation calculation module calculates the time and space correlation of each order in the list of orders to be delivered within a period of time, and groups the orders with correlation exceeding a threshold into a group, and jointly assigns a delivery person;

[0025] The delivery list generation module queries, for each group of orders or unbound single order, the delivery personnel whose real-time locations are within the set range of the starting point and the delivery personnel whose estimated locations are within the set range of the starting point at the start time, to form a list of optional delivery personnel;

[0026] The assignment module assigns couriers according to the couriers list until a couriers accepts the order.

[0027] Furthermore, the position estimation module includes a speed calculation unit, a path acquisition unit and a position calculation unit;

[0028] The speed calculation unit calculates the travel speed of the delivery person;

[0029] The route acquisition unit acquires the planned delivery route for the delivery person;

[0030] The position calculation unit calculates the position of the delivery person every k minutes according to the delivery path and the speed of the delivery person until the delivery person reaches the target position, thereby forming a trajectory over a period of time.

[0031] Furthermore, the association calculation module establishes a feature vector for clustering based on the attributes of the order, the attributes include: starting point location, target location, order item type and start time; for a cluster, the distance with other points is used as the association, and the orders corresponding to the points with an association less than a threshold are bundled. Furthermore, clustering is performed using K-MEANS, DBSCAN or BRICH model.

[0032] Furthermore, the dispatch list generation module includes a screening unit and a sorting unit;

[0033] The screening unit screens the couriers in the list of optional couriers, and removes couriers who do not accept orders, couriers who have rejected orders in the last 6 hours, couriers who have received a maximum number of orders, couriers who have reached a maximum capacity, blacklisted couriers, and couriers whose delivery types do not match;

[0034] The ranking unit calculates a comprehensive index for the selected delivery personnel and ranks them;

[0035] The comprehensive index calculation includes: Z = Q*k(P+M+R+H)

[0036] P is user satisfaction, Q is the deliveryman's order type, 1 is for one order and one delivery, -1 is for multiple orders and one delivery, M is the level score, k is the order type, 1 is for one order and one delivery, -1 is for multiple orders and one delivery, R is 0 for novice and 1 otherwise, and H is 5 for users who have collected Running Man.

[0037] Furthermore, the assignment module performs assignments according to a list of delivery personnel; after an order is assigned, the delivery personnel can choose to accept orders within a period of time, or accept part of a group of orders; for unaccepted orders, the next delivery personnel is assigned according to the list of optional delivery personnel. If there is no next delivery personnel, the returned order is set to the highest priority in the list of orders to be delivered and reassigned.

[0038] Furthermore, if the list of available delivery personnel is empty, the set range is expanded and the delivery personnel are searched again. If the list of available delivery personnel is still empty, the delivery personnel are added to the assigned list.

[0039] A module for grabbing orders is also provided, which adds a reward amount to the orders in the assignment list and then grabs the orders. If multiple people grab the order, the courier closest to the starting point will be selected; if no one grabs the order, the order will be dispatched manually.

[0040] The above technical solution of the present invention has the following beneficial technical effects:

[0041] (1) The present invention selects delivery personnel who are currently delivering goods based on their estimated locations, and selects delivery personnel who are not in the delivery state based on their real-time locations, thereby preventing delivery personnel from receiving orders that are not convenient for their route.

[0042] (2) The present invention associates orders, binds orders with an association degree exceeding a threshold into a group, and jointly assigns couriers, thereby saving manpower and material resources and improving the satisfaction of couriers.

[0043] (3) For orders with different starting and ending locations but on the way, since the time may be different, an order may have been delivered. By selecting the delivery person during delivery, the present invention can realize delivery by the same delivery person again.

[0044] (4) The present invention implements an order grabbing mechanism based on a reward mechanism for orders in areas with relatively few couriers or other difficult-to-deliver orders, thereby avoiding long delays for some orders. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a flowchart of the order assignment method;

[0046] Figure 2This is a schematic diagram of the order assignment system. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention is further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings. It should be understood that these descriptions are only exemplary and are not intended to limit the scope of the present invention. In addition, in the following description, the description of well-known structures and technologies is omitted to avoid unnecessary confusion of the concept of the present invention.

[0048] An order assignment method comprises the following steps:

[0049] (1) Detect the real-time location of the deliveryman who is in the process of delivering goods, and estimate the deliveryman's trajectory over a period of time based on the locations of the starting and ending points and the delivery time; detect the real-time location of the deliveryman who is not in the delivery state.

[0050] Estimating the delivery person's trajectory over a period of time, including: calculating the delivery person's speed, calculating the delivery person's position every k minutes according to the delivery path planned for the delivery person and the delivery person's speed, until the delivery person reaches the target location, forming a trajectory over a period of time. After the delivery person completes the delivery, the position is no longer estimated.

[0051] Only the location of the delivery personnel in the delivery process is estimated. Since the location of the delivery personnel in the delivery process changes rapidly and the behavior route is predictable, the dispatch is based on the expected location. For the delivery personnel who are not in the delivery state, the real-time location is detected and the dispatch is based on the real-time location.

[0052] (2) Obtain a list of orders to be delivered within a certain period of time.

[0053] For example, the order list includes orders to be delivered within 30 minutes, and the orders are arranged according to the delivery time. For returned orders or expedited orders, they are placed at the top of the order list.

[0054] Display pending orders, and search for orders based on the order city, service type, user type, whether it is the first order, and whether it is displayed on the open platform.

[0055] (3) Calculate the temporal and spatial correlation of each order in the list of orders to be delivered, group the orders whose correlation exceeds a threshold, and assign couriers to them together.

[0056] Factors that need to be considered in the correlation include: whether the type of order allows simultaneous assignment, the location coordinates of the order, the location coordinates of the deliveryman, the location of the ongoing orders on the deliveryman, the status of the ongoing orders on the deliveryman, the maximum number of orders the deliveryman can accept, the deliveryman's riding equipment, and other comprehensive correlations for calculation.

[0057] In one embodiment, orders of a type that does not allow simultaneous assignment (such as expedited orders) are eliminated, and feature vectors are established based on order attributes for clustering, the attributes include: starting location, target location, order item type, and start time. For a cluster, the distance to other points is used as the correlation, and the orders corresponding to points with a correlation less than a threshold are bundled. Furthermore, clustering is performed using K-MEANS, DBSCAN, or BRICH models.

[0058] (4) For each group of orders or unbound single order, query the delivery personnel whose real-time location is within the set range of the starting point and the delivery personnel whose estimated location at the start time is within the set range of the starting point to form a list of optional delivery personnel.

[0059] The attributes of the delivery person include: whether to accept orders, order rejection history, number of orders received in 24 hours, capacity utilization ratio, whether to be blacklisted, and delivery type.

[0060] The delivery type includes whether there is an insulated box and whether the delivery person is of a specific identity (whether it is a cake delivery person, a document delivery person, etc.).

[0061] The deliverymen in the list of available deliverymen are excluded: deliverymen who do not accept orders, deliverymen who have rejected orders in the last 6 hours, deliverymen who have received the maximum number of orders, deliverymen who have reached the maximum capacity, deliverymen who are blacklisted, and deliverymen who do not match the delivery type. The remaining deliverymen are ranked according to the comprehensive index.

[0062] The calculation of the deliveryman's comprehensive index includes: Z = Q*k(P+M+R+H)

[0063] P is user satisfaction, with a full score of 5 points; Q is the type of dispatch order, which is selected by the dispatcher. The dispatcher can choose to deliver multiple orders or one order at a time. One order and one delivery is 1, and multiple orders and one delivery is -1; M is the level score; k is the order type, which is selected by the user when placing an order. It can be delivered for one order or multiple orders. Multiple orders are -1 and 1 otherwise; R is 0 for a novice and 1 otherwise, and H is 5 otherwise 0 for a user's favorite running man. If the dispatcher type and the order type are consistent, the delivery can be made. Otherwise, the Z value is negative and the dispatcher is eliminated.

[0064] (5) Assign couriers according to the couriers list until a couriers accepts the order.

[0065] Furthermore, after the order is delivered, the deliveryman can choose to accept the order within a period of time, or accept part of the orders in a group of orders. If there is no response within the time limit, it will be deemed that the order is not accepted. For unaccepted orders, the next deliveryman will be assigned according to the list of optional deliverymen. If there is no next deliveryman, the returned order will be set to the highest priority in the list of orders to be delivered and reassigned.

[0066] If the list of available couriers is empty, expand the setting range and search for the couriers again. If the list of available couriers is still empty, add them to the assigned list. For example, if the original setting range is 1 km nearby, if there is no couriers, change it to 2 km nearby, and the maximum is 5 km nearby for searching. If you can't find the couriers, you can manually enter the couriers' phone number to assign them.

[0067] Furthermore, after adding a reward to the order in the assignment list, the order is grabbed. If multiple people grab the order, the courier closest to the starting point is selected; if no one grabs the order, the order is manually assigned. The backstage assigning personnel will artificially increase the order price by 1-10 yuan depending on the order timeout status and the order distance. The reward amount is, for example, a material reward or user level points. The higher the user level points, the higher the chance of getting the order.

[0068] On the other hand, an order assignment system is provided, combining Figure 2 , including real-time location detection module, location estimation module, correlation calculation module, delivery list generation module, assignment module and order grabbing module.

[0069] The real-time location detection module locates all delivery personnel in real time and detects their real-time locations.

[0070] The location estimation module estimates the trajectory of the deliveryman over a period of time based on the locations of the starting point and the end point, the real-time location, and the delivery time.

[0071] Furthermore, the position estimation module includes a speed calculation unit, a path acquisition unit and a position calculation unit; the speed calculation unit calculates the travel speed of the deliveryman; the path acquisition unit acquires the planned delivery path for the deliveryman; the position calculation unit calculates the position of the deliveryman every k minutes according to the delivery path and the speed of the deliveryman until the target position is reached, forming a trajectory over a period of time.

[0072] The correlation calculation module calculates the time and space correlation of each order in the list of orders to be delivered within a period of time, binds the orders with correlation exceeding a threshold into a group, and jointly assigns a delivery person.

[0073] Furthermore, the correlation calculation module establishes a feature vector based on the attributes of the order for clustering, and the attributes include: starting location, target location, order item type, and start time; for a cluster, the distance to other points is used as the correlation, and the orders corresponding to the points with a correlation less than a threshold are bundled. Furthermore, clustering is performed using K-MEANS, DBSCAN, or BRICH models.

[0074] The delivery list generation module queries, for each group of orders or unbound single order, the delivery persons whose real-time locations are within the set range of the starting point and the delivery persons whose estimated locations are within the set range of the starting point at the start time to form a list of optional delivery persons.

[0075] The delivery list generation module includes a screening unit and a sorting unit; the screening unit screens the couriers in the optional couriers list, and removes couriers who do not accept orders, couriers who have rejected orders in the past 6 hours, couriers who have reached the maximum number of orders, couriers who have reached the maximum transportation capacity, blacklisted couriers, and couriers whose delivery types do not match; the sorting unit calculates a comprehensive index for the screened couriers and ranks them.

[0076] The comprehensive index calculation includes: Z = Q*k(P+M+R+H)

[0077] P is user satisfaction; Q is the type of deliveryman dispatching orders, which the deliveryman can choose by himself. He can choose to deliver multiple orders or one order. One order and one delivery is 1, and multiple orders and one delivery is -1; M is the level points; k is the order type, which the user can choose when placing an order, whether to deliver one order or multiple orders, and multiple orders is -1 otherwise 1; R is 0 for novice otherwise 1, and H is 5 otherwise 0 for the user's favorite running man. If the calculation result Z is a negative value, it means that the type does not match, and the deliveryman is eliminated.

[0078] Select an order; load the order with nearby deliverymen, and arrange them from near to far according to the order's pickup point and the deliveryman's latest location. If there is no deliveryman within 1 km, you can change it to 2 km, with a maximum of 5 km. If you can't find the deliveryman, you can manually enter the deliveryman's phone number for assignment. Deliverymen and orders have types, and different types are incompatible.

[0079] The assignment module assigns couriers according to the couriers list until a couriers accepts the order.

[0080] The assignment module performs assignment according to the list of delivery personnel; after the order is assigned, the delivery personnel can choose to accept the order within a period of time, or accept part of the orders in a group of orders; for the unaccepted orders, the next delivery personnel is assigned according to the list of optional delivery personnel. If there is no next delivery personnel, the returned order is set to the highest priority in the list of orders to be delivered and reassigned.

[0081] Furthermore, if the list of available delivery personnel is empty, the setting range is expanded and the delivery personnel are searched again. If the list of available delivery personnel is still empty, the delivery personnel are added to the assigned list.

[0082] The order grabbing module adds a reward amount to the orders in the assignment list and then grabs the order. If multiple people grab the order, the courier closest to the starting point will be selected; if no one grabs the order, the order will be dispatched manually.

[0083] In summary, the present invention relates to an order assignment method and system, which detects the real-time position of the deliveryman in delivery, and estimates the trajectory of the deliveryman within a period of time according to the positions of the starting point and the end point and the delivery time; detects the real-time position of the deliveryman who is not in the delivery state; obtains a list of orders to be delivered within a period of time; calculates the time and space correlation of each order in the list of orders to be delivered, binds the orders with correlation exceeding the threshold into a group, and jointly assigns the deliveryman; for each group of orders or a single unbound order, queries the deliveryman whose real-time position is within the set range of the starting point and the deliveryman whose estimated position is within the set range of the starting point at the start time, and forms a list of optional deliverymen; assigns deliverymen according to the deliveryman list until a deliveryman accepts the order. The delivery method of the present invention makes order assignment more reasonable; similar orders are bundled to save manpower and material resources.

[0084] It should be understood that the above specific embodiments of the present invention are only used to illustrate or explain the principles of the present invention, and do not constitute a limitation of the present invention. Therefore, any modifications, equivalent substitutions, improvements, etc. made without departing from the spirit and scope of the present invention should be included in the protection scope of the present invention. In addition, the appended claims of the present invention are intended to cover all changes and modifications that fall within the scope and boundaries of the appended claims, or the equivalent forms of such scope and boundaries.

Claims

1. An order assignment method, It is characterized in that include: Detect the real-time location of the delivery person during delivery, and estimate the delivery person's trajectory over a period of time based on the locations of the starting and ending points and the delivery time; Check the real-time location of the delivery personnel who are not in the delivery state; Get a list of orders to be delivered within a certain period of time; Calculate the time and space correlation of each order in the list of orders to be delivered, bind the orders with correlation exceeding the threshold into a group, and assign delivery personnel together; The factors that need to be considered in the correlation include: whether the order type allows simultaneous assignment, the location coordinates of the order, the location coordinates of the deliveryman, the location of the order in progress on the deliveryman, the status of the order in progress on the deliveryman, the maximum number of orders the deliveryman can accept, and the deliveryman's riding equipment. For each group of orders or unbound single order, query the delivery personnel whose real-time location is within the set range of the starting point and the delivery personnel whose estimated location at the start time is within the set range of the starting point to form a list of optional delivery personnel; In the list of available couriers, exclude those who do not accept orders, those who have rejected orders in the last 6 hours, those who have received the maximum number of orders, those who have reached the maximum capacity, those who have been blacklisted, and those whose delivery types do not match. The remaining couriers are ranked according to the comprehensive index; The calculation of the deliveryman's comprehensive index includes: Z = Q*k*(P+M+R+H) P is user satisfaction, Q is the deliveryman's order type, 1 is one delivery for one order, -1 is one delivery for multiple orders, M is the level score, k is the order type, 1 is one delivery for one order, -1 is one delivery for multiple orders, R is 0 for novice and 1 otherwise, H is 5 for the user's favorite running man; Assign couriers according to the couriers list until a couriers accepts the order; The calculation of the temporal and spatial correlation of each order with other orders includes: removing orders of types that are not allowed to be assigned at the same time, establishing feature vectors for clustering based on the attributes of the orders, which include: starting location, target location, order item type, and start time. For a group, the distance to other points is used as the correlation, and the orders corresponding to the points with a correlation less than a threshold are bundled; Estimating the delivery person's trajectory over a period of time, including: calculating the delivery person's speed, calculating the delivery person's position every k minutes according to the planned delivery path and the delivery person's speed, until the delivery person reaches the target location, forming a trajectory over a period of time. After the delivery person completes the delivery, his position is no longer estimated.

2. The order assignment method according to claim 1, It is characterized in that Clustering was performed using K-MEANS, DBSCAN, or BRICH models.

3. The order assignment method according to claim 1, It is characterized in that If the list of available couriers is empty, expand the setting range and search for the couriers again. If the list of available couriers is still empty, add them to the assigned list. After adding the bonus amount to the order in the assignment list, the courier closest to the starting point will be selected if multiple people are competing for the order. If no one takes the order, it will be dispatched manually.

4. The order assignment method according to claim 3, It is characterized in that After the order is delivered, the courier can choose to accept orders within a period of time, or accept part of a group of orders; for unaccepted orders, the next courier will be assigned according to the list of available courier options. If there is no next courier, the returned order will be set to the highest priority in the list of orders to be delivered and reassigned.

5. An order assignment system, It is characterized in that It includes a real-time location detection module, a location estimation module, a correlation calculation module, a delivery list generation module, and an assignment module; The real-time location detection module locates all delivery personnel in real time and detects their real-time locations; The location estimation module estimates the trajectory of the delivery person over a period of time based on the locations of the starting point and the end point, the real-time location, and the delivery time. The correlation calculation module calculates the time and space correlation of each order in the list of orders to be delivered within a period of time, and groups the orders with correlation exceeding a threshold into a group, and jointly assigns a delivery person; The association calculation module establishes a feature vector for clustering based on the attributes of the order, and the attributes include: starting point location, target location, order item type and start time; for a group, the distance to other points is used as the association, and the orders corresponding to the points with an association less than a threshold are bundled; the factors to be considered for the association include: whether the type of order allows simultaneous assignment, the location coordinates of the order, the location coordinates of the deliveryman, the location of the order in progress on the deliveryman, the status of the order in progress on the deliveryman, the maximum number of orders accepted by the deliveryman, and the deliveryman's riding equipment are comprehensively associated for calculation; The delivery list generation module queries, for each group of orders or unbound single order, the delivery personnel whose real-time locations are within the set range of the starting point and the delivery personnel whose estimated locations are within the set range of the starting point at the start time, to form a list of optional delivery personnel; The delivery list generation module includes a screening unit and a sorting unit; The screening unit screens the couriers in the list of optional couriers, and removes couriers who do not accept orders, couriers who have rejected orders in the last 6 hours, couriers who have received a maximum number of orders, couriers who have reached a maximum capacity, blacklisted couriers, and couriers whose delivery types do not match; The ranking unit calculates a comprehensive index for the selected delivery personnel and ranks them; The comprehensive index calculation includes: Z = Q*k(P+M+R+H) P is user satisfaction, Q is the deliveryman's order type, 1 is one delivery for one order, -1 is one delivery for multiple orders, M is the level score, k is the order type, 1 is one delivery for one order, -1 is one delivery for multiple orders, R is 0 for novice and 1 otherwise, H is 5 for the user's favorite running man; The assignment module assigns couriers according to the couriers list until a couriers accepts the order; The position estimation module includes a speed calculation unit, a path acquisition unit and a position calculation unit; The speed calculation unit calculates the travel speed of the delivery person; The route acquisition unit acquires the planned delivery route for the delivery person; The position calculation unit calculates the position of the delivery person every k minutes according to the delivery path and the speed of the delivery person until the delivery person reaches the target position, thereby forming a trajectory over a period of time.

6. The order assignment system according to claim 5, It is characterized in that Clustering was performed using K-MEANS, DBSCAN, or BRICH models.

7. The order assignment system according to claim 5, It is characterized in that The assignment module performs assignment according to the list of delivery personnel; after the order is assigned, the delivery personnel can choose to accept the order within a period of time, or accept part of the orders in a group of orders; for the unaccepted orders, the next delivery personnel is assigned according to the list of optional delivery personnel. If there is no next delivery personnel, the returned order is set to the highest priority in the list of orders to be delivered and reassigned.

8. The order assignment system according to claim 7, It is characterized in that If the list of available couriers is empty, expand the setting range and search for the couriers again. If the list of available couriers is still empty, add them to the assigned list. A module for grabbing orders is also set up to grab orders after adding a bonus amount to the orders in the assignment list. If multiple people grab the order, the courier closest to the starting point will be selected; If no one takes the order, it will be dispatched manually.

Citation Information

Patent Citations

  • Dispatching method and apparatus, and server

    CN108520452A

  • An online order distribution and delivery method based on a scoring mechanism and route planning

    CN109002960A

  • Order distribution method, order distribution device, readable storage medium and electronic equipment

    CN109741142A

  • Order scheduling method and device, storage medium and electronic equipment

    CN113222305A