CN110264133B

CN110264133B - Vehicle allocation method and device, computer equipment and storage medium

Info

Publication number: CN110264133B
Application number: CN201910504784.4A
Authority: CN
Inventors: 谢恩宁; 刘玉春; 叶旺旺; 周星言
Original assignee: Zhejiang Dasou Vehicle Software Technology Co Ltd
Current assignee: Zhejiang Dasou Vehicle Software Technology Co Ltd
Priority date: 2019-06-12
Filing date: 2019-06-12
Publication date: 2022-04-19
Anticipated expiration: 2039-06-12
Also published as: CN110264133A

Abstract

The application relates to a vehicle deployment method, a vehicle deployment device, computer equipment and a storage medium. The method comprises the following steps: acquiring an unmatched vehicle purchasing order from an order pool according to a preset time period; extracting vehicle demands and order addresses in the unmatched vehicle purchasing orders; searching for available vehicles matched with the vehicle requirements from a vehicle pool, and acquiring the vehicle states and warehouse addresses of the available vehicles; inputting the warehouse address, the order address and the vehicle state into a time estimation model to obtain the waiting time of each saleable vehicle; the time estimation model completes order training according to history; processing the warehouse address, the order address and the waiting time to obtain a vehicle distribution scheme with minimum total waiting time; and scheduling and delivering the saleable vehicles according to a vehicle delivery scheme. By adopting the method, the time spent on vehicle delivery can be accurately estimated, so that the formulated vehicle delivery scheme is reasonable and reliable, and the reduction of the over-time rate of the whole order is realized.

Description

Vehicle allocation method and device, computer equipment and storage medium

Technical Field

The present application relates to the field of logistics technology, and in particular, to a vehicle deployment method, apparatus, computer device, and storage medium.

Background

After the customer signs an order for the vehicle with the vehicle seller, the vehicle seller needs to allocate the vehicle according to the order requirement. When the warehouses of the vehicle sellers are distributed in different cities, workers need to manually select and search the known warehouses, multi-region automatic matching cannot be completely realized, the allocation scheme obtained by the method lacks overall planning, resource mismatching is easily caused, and vehicle allocation efficiency is low.

Disclosure of Invention

In view of the above, there is a need to provide a vehicle dispatching method, device, computer device and storage medium, which can automatically dispatch vehicles according to order information and vendable vehicle warehousing addresses, thereby shortening the average vehicle dispatching time and improving the vehicle dispatching efficiency.

A vehicle deployment method, the method comprising:

acquiring an unmatched vehicle purchasing order from an order pool according to a preset time period;

extracting the vehicle demand and the order address in the unmatched vehicle purchasing order;

searching for available vehicles matched with the vehicle demands from a vehicle pool, and acquiring vehicle states and warehouse addresses of the available vehicles;

inputting the warehouse address, the order address and the vehicle state into a time estimation model to obtain the waiting time of each saleable vehicle; the time estimation model completes order training according to history;

and processing the warehouse address, the order address and the waiting time to obtain a vehicle distribution scheme with minimum total waiting time.

In one embodiment, before the obtaining the unmatched car purchase order from the order pool according to the preset time period, the method further includes:

receiving a vehicle purchasing order sent by a client terminal;

extracting vehicle demands and order addresses in the vehicle purchasing orders;

searching for a vehicle capable of selling the license plate in a vehicle pool of a first vehicle supply range corresponding to the order address, wherein the vehicle requirement is matched with the vehicle, and the vehicle state of the vehicle capable of selling the license plate is that a vehicle identification is associated with a license plate identification;

when the vehicle capable of being sold is not matched, storing the vehicle purchasing order as an unmatched vehicle purchasing order in an order pool;

and when the vehicle capable of selling the branded vehicles is matched, generating a vehicle distribution scheme according to the warehouse address of the vehicle capable of selling the branded vehicles and the order address of the order of the purchased vehicles.

In one embodiment, the processing the warehouse address, the order address, and the waiting time to obtain a vehicle delivery scenario with a minimum total waiting time includes:

and processing the warehouse address, the order address and the waiting time by adopting a bipartite graph weighted maximum matching algorithm or Hungarian algorithm to obtain a vehicle distribution scheme with the minimum total waiting time.

taking the warehouse address and the order address as nodes in a bipartite graph, and taking the waiting time between the nodes in the bipartite graph as a weight;

and obtaining the maximum weight matching of the bipartite graph to obtain a distribution scheme corresponding to the order of the available vehicle and the purchased vehicle with the minimum total waiting time.

In one embodiment, the method further comprises:

receiving saleable vehicle information sent by a warehousing management terminal, wherein the vehicle state in the saleable vehicle information is that a vehicle identifier is associated with a license plate identifier;

acquiring a second vehicle supply range corresponding to the saleable vehicle information according to the license plate identification;

matching the order address with the vehicle demand of the unmatched order corresponding to the second vehicle supply range with the available vehicle information;

when the matching of the available vehicle information and the unmatched vehicle purchasing orders fails, storing the available vehicle information into a vehicle pool;

and when the saleable vehicle information is successfully matched with the unmatched vehicle purchasing orders, generating a vehicle distribution scheme according to the warehouse address of the saleable vehicle information and the order address of the unmatched vehicle purchasing orders.

In one embodiment, the time estimation model is constructed in a manner including:

acquiring a preset number of historical completion orders as sample training orders;

extracting sample distribution information from the sample training order;

and performing machine learning training on the sample distribution information according to each sample training order to obtain a time estimation model.

In one embodiment, the machine learning training of the sample distribution information according to each sample training order to obtain the time estimation model further includes:

acquiring a historical completion order different from the sample training order as an evaluation verification order;

extracting historical distribution information from the evaluation verification order, wherein the historical distribution information comprises a historical warehouse address, a historical order address, historical waiting time and a historical vehicle state;

inputting the historical warehouse address, the historical order address and the historical vehicle state into the time estimation model to obtain verification estimated time;

and performing verification evaluation on the verification estimated time and the historical waiting time through evaluation standards, wherein the evaluation standards are at least one of the mean value of absolute deviation, the variance of absolute deviation, the median of absolute deviation and the R2 score.

In one embodiment, after the acquiring the vehicle status and the warehouse address of the saleable vehicle, the method further includes:

when the number of the available vehicles is judged to be smaller than the number of the unmatched vehicle purchasing orders, acquiring the current time and the order placing time in the vehicle purchasing orders;

ordering the vehicle purchasing orders according to the difference value between the current time and the ordering time;

inputting the warehouse address, the order address and the vehicle state in the vehicle purchasing orders with the difference arranged in the front row into a trained time estimation model according to the quantity of the available vehicles, wherein the vehicle purchasing orders are arranged from large to small according to the difference.

A vehicle blending apparatus, said apparatus comprising:

the vehicle purchasing order acquisition module is used for acquiring unmatched vehicle purchasing orders from the order pool according to a preset time period;

the order information extraction module is used for extracting the vehicle demand and the order address in the unmatched vehicle purchasing order;

the available vehicle searching module is used for searching available vehicles matched with the vehicle requirements from a vehicle pool and acquiring the vehicle states and warehouse addresses of the available vehicles;

the waiting time estimation module is used for inputting the warehouse address, the order address and the vehicle state into a time estimation model for finishing order training according to history to obtain the waiting time of each saleable vehicle;

and the distribution scheme generation module is used for processing the warehouse address, the order address and the waiting time to obtain a vehicle distribution scheme with the minimum total waiting time.

In one embodiment, the apparatus further comprises:

the order receiving module is used for receiving a vehicle purchasing order sent by the client terminal;

the order information extraction module is used for extracting the vehicle demand and the order address in the vehicle purchasing order;

the order matching module is used for searching for the vehicle capable of selling the license plate in a vehicle pool corresponding to the order address in the first vehicle supply range, wherein the vehicle requirement is matched with the vehicle, and the vehicle state of the vehicle capable of selling the license plate is that the vehicle identification is associated with the license plate identification;

the order storage module is used for storing the order of the purchased vehicle as an unmatched order of the purchased vehicle into an order pool when the sold vehicle is not matched;

and the vehicle allocation module is used for generating a vehicle distribution scheme according to the warehouse address of the marketable vehicle and the order address of the vehicle purchase order when the marketable vehicle is matched.

In another embodiment, the delivery plan generating module includes:

and the distribution scheme generating unit is used for processing the warehouse address, the order address and the waiting time by adopting a bipartite graph weighted maximum matching algorithm or a Hungarian algorithm to obtain a vehicle distribution scheme with the minimum total waiting time.

In some embodiments, the delivery plan generating module comprises:

a bipartite graph drawing unit, configured to use the warehouse address and the order address as nodes in a bipartite graph, and use the waiting time between the nodes in the bipartite graph as a weight;

and the scheme matching generation unit is used for solving the maximum weight matching of the bipartite graph to obtain a distribution scheme corresponding to the order of the available vehicle and the purchased vehicle, wherein the total waiting time is the minimum.

In one embodiment, the apparatus further comprises:

the system comprises a vehicle information receiving module, a storage management terminal and a vehicle information processing module, wherein the vehicle information receiving module is used for receiving saleable vehicle information sent by the storage management terminal, and the vehicle state in the saleable vehicle information is that a vehicle identifier is associated with a license plate identifier;

the supply range acquisition module is used for acquiring a second vehicle supply range corresponding to the saleable vehicle information according to the license plate identification;

the vehicle matching module is used for matching the vehicle demand of the unmatched order for the vehicle corresponding to the order address and the second vehicle supply range with the available vehicle information;

the vehicle storage module is used for storing the saleable vehicle information into a vehicle pool when the saleable vehicle information is failed to be matched with the unmatched vehicle purchasing order;

and the vehicle allocation module is used for generating a vehicle distribution scheme according to the warehouse address of the available vehicle information and the order address of the unmatched order when the available vehicle information and the unmatched order are successfully matched.

In one embodiment, the latency pre-estimation module comprises:

the training order obtaining unit is used for obtaining a preset number of historical completion orders as sample training orders;

the sample distribution information extraction unit is used for extracting sample distribution information from the sample training order;

and the learning training unit is used for performing machine learning training on the sample distribution information according to each sample training order to obtain a time estimation model.

In another embodiment, the latency pre-estimation module comprises:

an evaluation verification order obtaining unit configured to obtain a historical completion order different from the sample training order as an evaluation verification order;

a historical delivery information extraction unit, which is used for extracting historical delivery information from the evaluation verification order, wherein the historical delivery information comprises a historical warehouse address, a historical order address, historical waiting time and a historical vehicle state;

the waiting time estimation unit is used for inputting the historical warehouse address, the historical order address and the historical vehicle state into the time estimation model to obtain verification estimated time;

and the evaluation unit is used for carrying out verification evaluation on the verification estimated time and the historical waiting time through evaluation standards, wherein the evaluation standards are at least one of the mean value of absolute deviation, the variance of absolute deviation, the median of absolute deviation and the R2 score.

In some embodiments, the apparatus further comprises:

the time obtaining module is used for obtaining the current time and the order placing time in the order of the vehicle when the number of the available vehicles is judged to be smaller than the number of the unmatched order of the vehicle;

the order sorting module is used for sorting the order of the purchased vehicles according to the difference value between the current time and the order placing time;

and the information input module is used for inputting the warehouse address, the order address and the vehicle state in the vehicle purchasing orders with the difference arranged in the front row into a trained time estimation model according to the quantity of the available vehicles, and the vehicle purchasing orders are arranged according to the difference from large to small.

A computer device comprising a memory storing a computer program and a processor implementing the steps of the above method when executing the computer program.

A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned method.

According to the vehicle allocation method, the vehicle allocation device, the computer equipment and the storage medium, the unmatched vehicle purchasing orders are obtained from the order pool according to the preset time period; inputting the warehouse address, the order address and the vehicle state into a time estimation model which is trained according to the historical completed order, and obtaining the waiting time of each saleable vehicle; the warehouse address, the order address and the waiting time are processed to obtain a vehicle distribution scheme with the minimum total waiting time, a time estimation model is obtained by historical order completion, and the time spent on vehicle distribution is accurately estimated, so that the formulated vehicle distribution scheme is reasonable and reliable, the stable reduction of the over-period rate of the whole order is realized, and the standardization and the full automation of the whole matching process are realized. And the vehicle allocation method also establishes a three-dimensional and multi-dimensional order allocation system, reduces the delivery time of each order for the vehicle, reduces the overdue risk of the whole platform or company, reduces the total cost of vehicle transportation, and realizes the maximum overall benefit of the whole platform or company.

Drawings

FIG. 1 is a diagram illustrating an exemplary implementation of a vehicle deployment method;

FIG. 2 is a schematic flow chart diagram of a vehicle deployment method according to one embodiment;

FIG. 3 is a schematic flow chart diagram of a vehicle deployment method according to one embodiment;

FIG. 4 is a schematic flow chart diagram illustrating a vehicle dispatching method according to another embodiment;

FIG. 5 is a schematic diagram of a process for constructing a time estimation model according to another embodiment;

FIG. 6 is a schematic diagram of a process for constructing a time estimation model according to another embodiment;

FIG. 7 is a comparison chart of order delivery timeliness of the vehicle dispatching method in one embodiment;

FIG. 8 is a block diagram of a vehicle deployment apparatus in accordance with an exemplary embodiment;

FIG. 9 is a diagram illustrating an internal structure of a computer device according to an embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

The vehicle deployment method provided by the application can be applied to the application environment shown in fig. 1. The client terminal 102 communicates with the server 104 through a network, and the server 104 is in communication connection with the warehouse management terminal 106. The client terminal 102 sends the order for the vehicle to the server 104, and the server 104 stores the order for the vehicle in an order pool; the server 104 acquires the unmatched order for the vehicle from the order pool according to the preset time period; the server 104 extracts the vehicle demand and the order address in the unmatched vehicle purchasing order; the server 104 searches for available vehicles matching the vehicle requirements from the vehicle pool and acquires the vehicle states and warehouse addresses of the available vehicles; the server 104 inputs the warehouse address, the order address and the vehicle state into a time estimation model for completing order training according to history, and the waiting time of each saleable vehicle is obtained; the server 104 processes the warehouse address, the order address and the waiting time to obtain a vehicle distribution scheme with the minimum total waiting time; the server 104 schedules and dispatches the vendable vehicles according to a vehicle delivery schedule. The server 104 may send the vehicle distribution scheme to the warehouse management terminal 106 that manages the vendable vehicles, and the warehouse management terminal 106 deploys the vendable vehicles according to the vehicle distribution scheme. The client terminal 102 and the warehouse management terminal 106 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable smart devices, and the server 104 may be implemented by an independent server or a server cluster composed of a plurality of servers.

In one embodiment, as shown in fig. 2, a vehicle deployment method is provided, which is exemplified by the application of the method to the server 104 in fig. 1, and comprises the following steps:

step 202, obtaining the unmatched order for the vehicle from the order pool according to the preset time period.

The server 104 obtains the unmatched order for the vehicle from the order pool according to the preset time period. The predetermined time period may be set in advance by the vehicle seller managing the server or may be set by the system. The preset time period can be set to be 1-10 days, and can be adjusted according to order data. The order pool stores unmatched order orders for vehicles received by the vehicle seller from each client terminal 102, which are order orders for vehicles that are unmatched from the vehicle identification. The order for purchasing a vehicle at least comprises the vehicle requirement of the vehicle required by the customer and the order address of the final delivery of the vehicle, and the order for purchasing the vehicle also can comprise customer identification. The client identification can be identity information of the client and the like, and has uniqueness.

And step 204, extracting the vehicle demand and the order address in the unmatched vehicle purchasing order.

The server 104 extracts the vehicle demand and the order address in the unmatched purchase order. The vehicle requirements may be various configuration information of the vehicle desired by the customer, such as vehicle model, color, engine model, etc. The order address is the address at which the vehicle is ultimately shipped, and may be a customer-designated sales store or warehouse or other location.

And step 206, searching for a saleable vehicle matched with the vehicle requirement from the vehicle pool, and acquiring the vehicle state and the warehouse address of the saleable vehicle.

The vehicle pool stores vendable vehicle information held by the vehicle seller that has not yet matched the customer order. The vehicle information includes a vehicle identification and a vehicle configuration corresponding to the vehicle identification. The vehicle identification is a unique identification that identifies each vendable vehicle, and may be a vehicle identification code (VIN code). The vehicle identification code is determined according to the national vehicle management standard and includes information on the manufacturer, the year, the model, the body type and code, the engine code, the assembly site, and the like of the vehicle. The vehicle configuration may be body color, wheelbase, engine power, etc.

The vehicle status has a plurality of types, and each vehicle corresponds to one vehicle status, but the vehicle status can be updated in real time according to the sales status of the available vehicles. For example, the vehicle status may be present and the vehicle identification is associated with the license plate identification, present and the vehicle identification is not associated with the license plate identification but has been invoiced to the customer, in-transit with the vehicle identification not associated with the license plate identification, in-purchase with the vehicle identification not associated with the license plate identification, and the like. The warehouse address is the address of the warehouse where the vendable vehicle is stored.

The server 104 searches for a vendable vehicle from the pool of vehicles that matches the vehicle demand and obtains the vehicle status and the warehouse address of the vendable vehicle. The server can obtain the vehicle configuration information according to the VIN code, match the vehicle configuration information with the vehicle requirements of the order, and obtain the vehicle state and the warehouse address of the saleable vehicle according to the VIN code after matching is successful. The server can search at least one available vehicle matched with the vehicle requirement from the vehicle pool according to one order, and each available vehicle information can be matched with at least one order. When the vehicle configuration information contains the vehicle requirement, the server judges that the available vehicle is successfully matched with the vehicle requirement; when the vehicle configuration information does not include the vehicle demand, the server determines that the vendible vehicle fails to match the vehicle demand.

Step 208, inputting the warehouse address, the order address and the vehicle state into a time estimation model to obtain the waiting time of each saleable vehicle; the time estimation model is trained according to the history completed order.

The server 104 inputs the warehouse address, the order address and the vehicle state into a time estimation model for completing order training according to history, and the waiting time of each saleable vehicle is obtained. The time estimation model is obtained by training according to historical order addresses, historical warehouse addresses, historical waiting time and historical vehicle states in historical finished orders. The waiting time obtained by the time estimation model comprises distribution time, procedure time and the like. The delivery time is derived from the distance and path between the order address and the warehouse address. The procedure time refers to the time taken to complete the procedure of transferring ownership of the vehicle, and the procedure time taken for different vehicle states and different order addresses is different.

For example, when an unmatched order for a vehicle matches a plurality of vendable vehicles, the server calculates the wait time for each vendable vehicle to be assigned to the order address of the unmatched order for the vehicle. When a plurality of unmatched shopping vehicle orders are matched with a plurality of saleable vehicles, the server respectively calculates the waiting time of each saleable vehicle for being distributed to the order address of each unmatched shopping vehicle order. When there is a vendable vehicle matching a plurality of unmatched order for the vehicle, the server calculates a wait time for the vendable vehicle to be delivered to the order address of each unmatched order for the vehicle.

And step 210, processing the warehouse address, the order address and the waiting time to obtain a vehicle distribution scheme with minimum total waiting time.

The server 104 processes the warehouse address, the order address, and the waiting time to obtain a vehicle delivery plan with the minimum total waiting time. The server 104 performs bipartite graph processing on the warehouse address, the order address and the waiting time, and obtains a vehicle distribution scheme with the minimum total waiting time according to a bipartite graph algorithm. The vehicle delivery scheme associates the vehicle identification with the order for the vehicle purchase and generates a delivery path based on the warehouse address and the order address.

In one embodiment, processing the warehouse address, the order address, and the wait time to obtain a vehicle delivery scenario with a minimum total wait time includes: and processing the warehouse address, the order address and the waiting time by adopting a bipartite graph weighted maximum matching algorithm or Hungarian algorithm to obtain a vehicle distribution scheme with the minimum total waiting time.

The server 104 may process the warehouse address, the order address, and the wait time using a bipartite graph algorithm to obtain a vehicle delivery scenario with a minimum total wait time. The bipartite graph algorithm can be a bipartite graph weighted maximum matching algorithm or a hungarian algorithm or the like.

The server 104 schedules and dispatches the vendable vehicles according to a vehicle delivery schedule. The server 104 may send the vehicle distribution scheme to the warehouse management terminal 106 that manages the vendable vehicles, and the warehouse management terminal 106 deploys the vendable vehicles according to the vehicle distribution scheme.

According to the vehicle allocation method, the unmatched vehicle purchasing orders are obtained from the order pool according to the preset time period; inputting the warehouse address, the order address and the vehicle state into a time estimation model which is trained according to the historical completed order, and obtaining the waiting time of each saleable vehicle; the warehouse address, the order address and the waiting time are processed to obtain a vehicle distribution scheme with the minimum total waiting time, a time estimation model is obtained by historical order completion, and the time spent on vehicle distribution is accurately estimated, so that the formulated vehicle distribution scheme is reasonable and reliable, the stable reduction of the over-period rate of the whole order is realized, and the standardization and the full automation of the whole matching process are realized. And the vehicle allocation method also establishes a three-dimensional and multi-dimensional order allocation system, reduces the delivery time of each order for the vehicle, reduces the overdue risk of the whole platform or company, reduces the total cost of vehicle transportation, and realizes the maximum overall benefit of the whole platform or company.

In one embodiment, as shown in fig. 3, before obtaining the unmatched order for the vehicle from the order pool according to the preset time period, there are the following steps:

step 302, receiving a vehicle purchase order sent by the client terminal.

The server 104 receives the order for the vehicle sent by the client terminal 102. The client terminal 102 may generate a vehicle order based on the vehicle requirements and delivery address of the desired vehicle entered by the client and then send the order to the server 104.

And step 304, extracting the vehicle demand and the order address in the vehicle purchasing order.

The server 104 extracts the vehicle demand and the order address in the order for the vehicle.

Step 306, searching for a vehicle capable of selling the license plate which is matched with the vehicle requirement in a vehicle pool of the first vehicle supply range corresponding to the order address, wherein the vehicle state of the vehicle capable of selling the license plate is that the vehicle identification is associated with the license plate identification.

The marketable card vehicle is a card vehicle held by a vehicle seller and that has not yet matched a customer order. The vehicle state of the marketable vehicle is that the vehicle identification is associated with the license plate identification. The license plate identification may be a license plate number. When the vehicle identifier is associated with the license plate identifier, the first vehicle supply range is reduced to an area corresponding to the license plate identifier. To avoid inventory, there is a need to match the marketable vehicles in a timely manner. The server 104 searches for a vehicle pool in the first vehicle supply range corresponding to the order address for a marketable vehicle that matches the vehicle demand.

And 308, when the vehicle capable of being sold is not matched, storing the order for purchasing the vehicle as an unmatched order for purchasing the vehicle in an order pool.

When the server 104 does not match a marketable vehicle, the server 104 stores the order for the purchased vehicle as an unmatched order for the purchased vehicle in the order pool and waits for the next match.

And 310, when the vehicle capable of selling the cards is matched, generating a vehicle distribution scheme according to the warehouse address of the vehicle capable of selling the cards and the order address of the order of the purchased vehicle.

When the server 104 matches the available vehicles, the server 104 generates a vehicle delivery plan according to the warehouse address of the available vehicles and the order address of the order of the purchased vehicles, and schedules and delivers the available vehicles according to the vehicle delivery plan. The server 104 may also send the vehicle distribution scheme to the warehouse management terminal 106 that manages the saleable vehicles, and the warehouse management terminal 106 schedules and distributes the saleable vehicles according to the vehicle distribution scheme.

According to the vehicle allocation method, when the server receives the vehicle purchasing order sent by the client terminal, the server firstly matches the vehicle purchasing order in the vehicle pool corresponding to the order address in the first vehicle supply range, so that the inventory time of the vehicle capable of selling the branded vehicle is reduced, and the total cost of vehicle management is reduced.

In one embodiment, the warehouse address, the order address and the waiting time are processed to obtain a vehicle distribution scheme with minimum total waiting time, and the method comprises the following steps: taking the warehouse address and the order address as nodes in a bipartite graph, and taking the waiting time between the nodes in the bipartite graph as a weight; and obtaining the maximum weight matching of the bipartite graph to obtain a distribution scheme corresponding to the order of the available vehicle and the purchased vehicle with the minimum total waiting time.

The server 104 obtains the vehicle distribution scheme by using a bipartite graph weighted maximum matching algorithm, which may be a Kuhn-Munkres algorithm or the like. The server 104 takes the warehouse address and the order address as nodes in the bipartite graph and the wait time between the nodes in the bipartite graph as a weight. The server may construct a matrix of the bipartite graph with the warehouse address as node X in the bipartite graph, the order address as node Y in the bipartite graph, and the wait time as a weight.

The server 104 can obtain the distribution scheme corresponding to the order of the available vehicle and the purchased vehicle with the minimum total waiting time by obtaining the maximum weight matching of the bipartite graph according to the matrix of the bipartite graph; the server 104 may also traverse the bipartite graph. The match in a bipartite graph is a set of edges that have no common nodes. If the weight sum of each edge of one matching is the maximum in all the matches of one bipartite graph, the matching is the maximum weight matching. The server may traverse all combinations of edges in the bipartite graph to find the maximum weight match. After the maximum weight match is obtained, the warehouse address and the order address associated with the edge in the maximum weight match can generate a distribution scheme corresponding to the available vehicle and the order of the purchased vehicle with the minimum total waiting time.

According to the vehicle allocation method, the server adopts the bipartite graph weighted maximum matching algorithm to accurately obtain the distribution scheme corresponding to the available vehicle with the minimum total waiting time and the vehicle purchasing order, so that the average delivery time of each vehicle purchasing order is further reduced, and the overdue risk of the whole platform or company is reduced.

In one embodiment, as shown in fig. 4, the vehicle deployment method further comprises the steps of:

step 402, receiving saleable vehicle information sent by the warehousing management terminal, wherein the vehicle state in the saleable vehicle information is that the vehicle identification is associated with the license plate identification.

The server 104 receives the saleable vehicle information sent by the warehousing management terminal 106, and the vehicle state in the saleable vehicle information is that the vehicle identification is associated with the license plate identification. The warehousing management terminal 106 may update the various vehicle information entered into the vehicle vendor's possession in real-time.

And step 404, acquiring a second vehicle supply range corresponding to the saleable vehicle information according to the license plate identifier.

And the server 104 acquires a second vehicle supply range corresponding to the saleable vehicle information according to the license plate identification. The license plate identification corresponds to a province or a city, and when the vehicle identification is associated with the license plate identification, the second vehicle supply range is the city or the province corresponding to the license plate identification.

And 406, matching the vehicle demand of the unmatched order corresponding to the order address and the second vehicle supply range with the available vehicle information.

The server 104 matches the vehicle demand of the unmatched order for the vehicle corresponding to the order address and the second vehicle supply range with the available vehicle information. When the order address includes the second vehicle supply range, the server determines that the order address corresponds to the second vehicle supply range, for example, the second vehicle supply range is hangzhou city, the order address is XX road in xiaoshan district of hangzhou city, the order address includes the second vehicle supply range, and the server determines that the order address corresponds to the second vehicle supply range. When the available vehicle information contains the vehicle requirement, the server judges that the available vehicle information is successfully matched with the unmatched vehicle purchasing order; when the available vehicle information does not contain the vehicle requirement, the server judges that the available vehicle information fails to be matched with the unmatched vehicle purchasing order.

And step 408, when the matching of the available vehicle information and the unmatched vehicle purchasing order fails, storing the available vehicle information into a vehicle pool.

When the vendable vehicle information fails to match an unmatched purchase order, server 104 stores the vendable vehicle information in a pool of vehicles.

And step 410, when the saleable vehicle information is successfully matched with the unmatched vehicle purchasing orders, generating a vehicle distribution scheme according to the warehouse address of the saleable vehicle information and the order address of the unmatched vehicle purchasing orders.

When the available vehicle information is successfully matched with the unmatched order for the vehicle, the server 104 generates a vehicle distribution scheme according to the warehouse address of the available vehicle information and the order address of the unmatched order for the vehicle, and schedules and distributes the vehicle on sale according to the vehicle distribution scheme. Each vendable vehicle information may be matched to an order.

According to the vehicle allocation method, when the server receives the information of the available vehicles sent by the warehousing management terminal and the available vehicles are the vehicles on the license plate, the server firstly determines the second vehicle supply range of the available vehicles and timely matches with the order of the purchased vehicles with the order addresses in the second vehicle supply range, so that the inventory time of the available vehicles on the license plate is reduced, and the total cost of vehicle management is reduced.

In one embodiment, as shown in fig. 5, the time estimation model is constructed by the following steps:

step 502, obtaining a preset number of historical completion orders as sample training orders.

Server 104 obtains a preset number of historical completion orders as sample training orders. The historical completion order is a historical order of each vehicle sold by the vehicle seller, and at least comprises a historical warehouse address of the historical vehicle, a historical order address, a historical waiting time and a historical vehicle state. The preset number can be set according to the number of historical completed orders or can be set by default in the system. The predetermined amount may be a numerical value or a percentage of the number of historical completed orders.

Step 504, sample distribution information is extracted from the sample training order.

The server 104 extracts sample delivery information from the sample training order, the sample delivery information including factors that affect vehicle delivery latency. The sample delivery information may include a sample warehouse address, a sample order address, a sample wait time, and a sample vehicle status.

Step 506, performing machine learning training on the sample distribution information according to each sample training order to obtain a time estimation model.

The server 104 performs machine learning training on the sample distribution information according to each sample training order to obtain a time estimation model. The server may derive a location-distance-delivery time association based on each sample warehouse address, sample order address, and delivery date, and a status-address-procedure time association based on each vehicle status, order address, sample wait time, and sample delivery time. The server establishes a time estimation model according to the association between the position-distance-distribution time and the association between the state-address-procedure time.

According to the vehicle allocation method, the vehicle distribution time and the procedure time required by vehicle transfer are considered, and the waiting time is accurately estimated.

In an embodiment, as shown in fig. 6, the machine learning training is performed on the sample distribution information according to each sample training order to obtain the time estimation model, and the method further includes the following steps:

step 602, obtaining a historical completion order different from the sample training order as an evaluation validation order.

Server 104 obtains a historical completion order that is different from the sample training order as an evaluation validation order. The evaluation validation order and the sample training orders both belong to historical completion orders, but the evaluation validation order is different from each of the historical completion orders in the sample training orders. The number of the evaluation verification orders and the number of the sample training orders can be set in proportion, and the system can also set according to the accuracy.

Step 604, extracting historical delivery information from the evaluation verification order, wherein the historical delivery information comprises historical warehouse addresses, historical order addresses, historical waiting time and historical vehicle states.

The server 104 extracts historical delivery information from the evaluation validation order, which may include historical warehouse addresses, historical order addresses, historical wait times, and historical vehicle status. The historical shipping information may also include historical shipping dates and other factors that affect the wait time.

And 606, inputting the historical warehouse address, the historical order address and the historical vehicle state into the time estimation model to obtain verification estimated time.

The server 104 inputs the historical warehouse address, the historical order address and the historical vehicle state into the time estimation model to obtain the verification estimated time.

And 608, performing verification evaluation on the verification estimated time and the historical waiting time through evaluation standards, wherein the evaluation standards are at least one of a mean value of absolute deviation, a variance of absolute deviation, a median of absolute deviation and a R2 score.

The server 104 performs verification evaluation on the verification prediction time and the historical waiting time through evaluation criteria, wherein the evaluation criteria are at least one of a mean value of absolute deviation, a variance of absolute deviation, a median of absolute deviation and an R2 score. The server calculates the time difference between the verification estimated time and the historical waiting time, calculates the absolute deviation between the time differences, and substitutes the difference and the absolute deviation into a calculation formula according to the evaluation standard to obtain the accuracy of the time estimation model. And when the accuracy is greater than the preset threshold, the time estimation model is verified to be qualified, and corresponding model parameters are repeatedly adjusted until the model can meet corresponding standards.

Mean of absolute deviation:

variance of absolute deviation:

median absolute deviation:

r2 score:

wherein

Wherein n is_samplesQuantity, y, representing all assessment validation orders_iAnd

respectively representing the historical waiting time of each evaluation verification order and the verification estimated time calculated by the time estimation model.

The model is verified and evaluated from different angles respectively by the different evaluation standards, and the confidence degree of the model is described. The mean value of the absolute deviation reflects the overall deviation degree, but the evaluation of abnormal data is deviated; the variance of the absolute deviation is mainly that in the case of equivalent means, a lower variance may indicate a smaller estimation error; the median of absolute deviations accounts for the deviation in most cases, but the overall deviation may be large; the R2 score, also called goodness of fit, reflects the difference between the variance of the predicted values and the variance of the true values, with closer to 1 indicating that the overall distribution from which the classification data came is consistent with the predicted distribution.

In another embodiment, after the obtaining the vehicle status and the warehouse address of the saleable vehicle, the method further comprises the following steps: when the number of the available vehicles is judged to be smaller than the number of the unmatched vehicle purchasing orders, acquiring the current time and the order placing time in the vehicle purchasing orders; ordering the vehicle purchasing orders according to the difference value between the current time and the ordering time; inputting the warehouse address, the order address and the vehicle state in the vehicle purchasing orders with the difference arranged in the front row into a trained time estimation model according to the quantity of the available vehicles, wherein the vehicle purchasing orders are arranged from large to small according to the difference.

Server 104 determines whether the number of available vehicles is greater than the number of unmatched purchase orders. When the server judges that the number of the available vehicles is larger than the number of the unmatched vehicle purchasing orders, the server calculates the waiting time corresponding to each available vehicle; when the server 104 determines that the number of available vehicles is less than the number of unmatched purchase orders, the server 104 obtains the current time and the order placement time in the purchase order. And the server 104 sorts the order of the vehicle purchase according to the difference value between the current time and the order placing time to obtain an order sorting result. The server 104 extracts the vehicle-purchasing orders with the difference in the order sorting result ranked in the front according to the number of the available vehicles, and the vehicle-purchasing orders in the order sorting result are ranked from large to small according to the difference. And the server inputs the extracted warehouse address, the order address and the vehicle state which are not matched with the vehicle purchasing order into the trained time estimation model.

According to the vehicle allocation method, the order distribution is adjusted in real time according to the conditions of the available vehicles, the overtime risk is prevented from being increased due to long-term overstock of partial orders, and a small number of available vehicles are preferentially matched with the orders to be overtime, so that the minimum overtime rate of the whole system is guaranteed to a relatively reasonable degree.

As shown in fig. 7, the order delivery age of the history matching result is: it takes 11.54 days to complete 25% of orders, 14.39 days to complete 50% of orders, an average of 18.24 days to complete 75% of orders, and 44.53 days to maximum completion time. The order delivery timeliness of the model matching result in the embodiment is as follows: it takes 10.96 days to complete 25% of the orders, 13.18 days to complete 50% of the orders, an average of 15.99 days to complete 75% of the orders, and 35.68 days for maximum completion time. The vehicle allocation method enables the whole order overrun rate to be reduced by 20%, and the whole delivery time length to be shortened by 25% to the maximum extent.

It should be understood that although the various steps in the flow charts of fig. 2-6 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2-6 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performance of the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternating with other steps or at least some of the sub-steps or stages of other steps.

In one embodiment, as shown in fig. 8, there is provided a vehicle deployment apparatus comprising: a vehicle purchase order acquisition module 802, an order information extraction module 804, a vendable vehicle search module 806, a wait time estimation module 808, and a delivery plan generation module 810, wherein:

a vehicle purchase order obtaining module 802, configured to obtain an unmatched vehicle purchase order from the order pool according to a preset time period.

And the order information extraction module 804 is used for extracting the vehicle demand and the order address in the unmatched vehicle purchasing order.

A vendable vehicle search module 806 that searches a pool of vehicles for a vendable vehicle that matches the vehicle demand and obtains a vehicle status and a warehouse address for the vendable vehicle.

The waiting time estimation module 808 is configured to input the warehouse address, the order address, and the vehicle state into a time estimation model to obtain waiting time of each saleable vehicle; the time estimation model is trained according to the history completed order.

And the distribution scheme generating module 810 is configured to process the warehouse address, the order address, and the waiting time to obtain a vehicle distribution scheme with the minimum total waiting time.

In some embodiments, the apparatus further comprises an order receiving module, an order information extraction module, an order matching module, an order storage module, and a vehicle deployment module, wherein:

and the order receiving module is used for receiving the vehicle purchasing order sent by the client terminal.

And the order information extraction module is used for extracting the vehicle demand and the order address in the vehicle purchasing order.

And the order matching module is used for searching for the vehicle capable of selling the license plate matched with the vehicle requirement in a vehicle pool of the first vehicle supply range corresponding to the order address, and the vehicle state of the vehicle capable of selling the license plate is that the vehicle identification is associated with the license plate identification.

And the order storage module is used for storing the order of the purchased vehicle as an unmatched order of the purchased vehicle into an order pool when the marketable vehicle is not matched.

In some embodiments, the delivery plan generating module 810 includes a delivery plan generating unit, wherein:

In another embodiment, the distribution scheme generating module 810 includes a bipartite graph drawing unit and a scheme matching generating unit, wherein:

and the bipartite graph drawing unit is used for taking the warehouse address and the order address as nodes in the bipartite graph and taking the waiting time between the nodes in the bipartite graph as a weight.

In one embodiment, the apparatus further comprises a vehicle information receiving module, a supply range obtaining module, a vehicle matching module, a vehicle storage module, and a vehicle allocation module, wherein:

the vehicle information receiving module is used for receiving the saleable vehicle information sent by the warehousing management terminal, and the vehicle state in the saleable vehicle information is that the vehicle identification is associated with the license plate identification.

And the supply range acquisition module is used for acquiring a second vehicle supply range corresponding to the saleable vehicle information according to the license plate identification.

And the vehicle matching module is used for matching the vehicle demand of the unmatched order for the vehicle corresponding to the order address and the second vehicle supply range with the available vehicle information.

And the vehicle storage module is used for storing the saleable vehicle information into a vehicle pool when the saleable vehicle information is failed to be matched with the unmatched vehicle purchasing order.

In one embodiment, the latency estimator module 808 includes a training order acquisition unit, a sample distribution information extraction unit, and a learning training unit, wherein:

and the training order obtaining unit is used for obtaining a preset number of historical completion orders as sample training orders.

And the sample distribution information extraction unit is used for extracting sample distribution information from the sample training order.

In one embodiment, the waiting time estimation module 808 further comprises an evaluation verification order obtaining unit, a historical delivery information extracting unit, a waiting time estimation unit and an evaluation unit, wherein:

and the evaluation verification order acquisition unit is used for acquiring a historical completion order different from the sample training order as an evaluation verification order.

And the historical delivery information extraction unit is used for extracting historical delivery information from the evaluation verification order, and the historical delivery information comprises a historical warehouse address, a historical order address, a historical waiting time and a historical vehicle state.

And the waiting time estimation unit is used for inputting the historical warehouse address, the historical order address and the historical vehicle state into the time estimation model to obtain the verification estimated time.

In one embodiment, the apparatus further comprises a time acquisition module, an order ordering module, and an information input module, wherein:

For specific limitations of the vehicle deployment apparatus, reference may be made to the above limitations of the vehicle deployment method, which are not described herein again. The various modules in the vehicle configuration apparatus described above may be implemented in whole or in part by software, hardware, and combinations thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 9. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used to store vehicle pool data and order pool data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a vehicle deployment method.

Those skilled in the art will appreciate that the architecture shown in fig. 9 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, there is provided a computer device comprising a memory storing a computer program and a processor implementing the following steps when the processor executes the computer program:

In one embodiment, the processor, when executing the computer program, further performs the steps of obtaining the unmatched order for the vehicle from the order pool according to a preset time period, before: receiving a vehicle purchasing order sent by a client terminal; extracting vehicle demands and order addresses in the vehicle purchasing orders; searching for a vehicle capable of selling the license plate in a vehicle pool of a first vehicle supply range corresponding to the order address, wherein the vehicle requirement is matched with the vehicle, and the vehicle state of the vehicle capable of selling the license plate is that a vehicle identification is associated with a license plate identification; when the vehicle capable of being sold is not matched, storing the vehicle purchasing order as an unmatched vehicle purchasing order in an order pool; and when the vehicle capable of selling the branded vehicles is matched, generating a vehicle distribution scheme according to the warehouse address of the vehicle capable of selling the branded vehicles and the order address of the order of the purchased vehicles.

In one embodiment, the processor when executing the computer program performs the step of processing the warehouse address, the order address, and the wait time to obtain a vehicle delivery schedule with a minimum total wait time, further configured to: and processing the warehouse address, the order address and the waiting time by adopting a bipartite graph weighted maximum matching algorithm or Hungarian algorithm to obtain a vehicle distribution scheme with the minimum total waiting time.

In one embodiment, the processor when executing the computer program performs the step of processing the warehouse address, the order address, and the wait time to obtain a vehicle delivery schedule with a minimum total wait time, further configured to: taking the warehouse address and the order address as nodes in a bipartite graph, and taking the waiting time between the nodes in the bipartite graph as a weight; and obtaining the maximum weight matching of the bipartite graph to obtain a distribution scheme corresponding to the order of the available vehicle and the purchased vehicle with the minimum total waiting time.

In one embodiment, the processor, when executing the computer program, further performs the steps of: receiving saleable vehicle information sent by a warehousing management terminal, wherein the vehicle state in the saleable vehicle information is that a vehicle identifier is associated with a license plate identifier; acquiring a second vehicle supply range corresponding to the saleable vehicle information according to the license plate identification; matching the order address with the vehicle demand of the unmatched order corresponding to the second vehicle supply range with the available vehicle information; when the matching of the available vehicle information and the unmatched vehicle purchasing orders fails, storing the available vehicle information into a vehicle pool; and when the saleable vehicle information is successfully matched with the unmatched vehicle purchasing orders, generating a vehicle distribution scheme according to the warehouse address of the saleable vehicle information and the order address of the unmatched vehicle purchasing orders.

In one embodiment, the processor, when executing the computer program, further performs the step of constructing the time estimation model by: acquiring a preset number of historical completion orders as sample training orders; extracting sample distribution information from the sample training order; and performing machine learning training on the sample distribution information according to each sample training order to obtain a time estimation model.

In one embodiment, the processor, when executing the computer program, performs machine learning training on the sample distribution information according to each sample training order to obtain the time estimation model, is further configured to: acquiring a historical completion order different from the sample training order as an evaluation verification order; extracting historical distribution information from the evaluation verification order, wherein the historical distribution information comprises a historical warehouse address, a historical order address, historical waiting time and a historical vehicle state; inputting the historical warehouse address, the historical order address and the historical vehicle state into the time estimation model to obtain verification estimated time; and performing verification evaluation on the verification estimated time and the historical waiting time through evaluation standards, wherein the evaluation standards are at least one of the mean value of absolute deviation, the variance of absolute deviation, the median of absolute deviation and the R2 score.

In one embodiment, the processor, when executing the computer program, further performs the steps of obtaining the vehicle status and the warehouse address of the vendible vehicle, and further: when the number of the available vehicles is judged to be smaller than the number of the unmatched vehicle purchasing orders, acquiring the current time and the order placing time in the vehicle purchasing orders; ordering the vehicle purchasing orders according to the difference value between the current time and the ordering time; inputting the warehouse address, the order address and the vehicle state in the vehicle purchasing orders with the difference arranged in the front row into a trained time estimation model according to the quantity of the available vehicles, wherein the vehicle purchasing orders are arranged from large to small according to the difference.

In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:

In one embodiment, the computer program when executed by the processor further performs the steps of: receiving a vehicle purchasing order sent by a client terminal; extracting vehicle demands and order addresses in the vehicle purchasing orders; searching for a vehicle capable of selling the license plate in a vehicle pool of a first vehicle supply range corresponding to the order address, wherein the vehicle requirement is matched with the vehicle, and the vehicle state of the vehicle capable of selling the license plate is that a vehicle identification is associated with a license plate identification; when the vehicle capable of being sold is not matched, storing the vehicle purchasing order as an unmatched vehicle purchasing order in an order pool; and when the vehicle capable of selling the branded vehicles is matched, generating a vehicle distribution scheme according to the warehouse address of the vehicle capable of selling the branded vehicles and the order address of the order of the purchased vehicles.

In one embodiment, the computer program when executed by the processor performs the step of processing the warehouse address, the order address, and the wait time to obtain a vehicle delivery schedule with a minimum total wait time further operable to: and processing the warehouse address, the order address and the waiting time by adopting a bipartite graph weighted maximum matching algorithm or Hungarian algorithm to obtain a vehicle distribution scheme with the minimum total waiting time.

In one embodiment, the computer program when executed by the processor performs the step of processing the warehouse address, the order address, and the wait time to obtain a vehicle delivery schedule with a minimum total wait time further operable to: taking the warehouse address and the order address as nodes in a bipartite graph, and taking the waiting time between the nodes in the bipartite graph as a weight; and obtaining the maximum weight matching of the bipartite graph to obtain a distribution scheme corresponding to the order of the available vehicle and the purchased vehicle with the minimum total waiting time.

In one embodiment, the computer program when executed by the processor is further operable to: receiving saleable vehicle information sent by a warehousing management terminal, wherein the vehicle state in the saleable vehicle information is that a vehicle identifier is associated with a license plate identifier; acquiring a second vehicle supply range corresponding to the saleable vehicle information according to the license plate identification; matching the order address with the vehicle demand of the unmatched order corresponding to the second vehicle supply range with the available vehicle information; when the matching of the available vehicle information and the unmatched vehicle purchasing orders fails, storing the available vehicle information into a vehicle pool; and when the saleable vehicle information is successfully matched with the unmatched vehicle purchasing orders, generating a vehicle distribution scheme according to the warehouse address of the saleable vehicle information and the order address of the unmatched vehicle purchasing orders.

In one embodiment, the computer program when executed by the processor further performs the step of constructing the time prediction model by: acquiring a preset number of historical completion orders as sample training orders; extracting sample distribution information from the sample training order; and performing machine learning training on the sample distribution information according to each sample training order to obtain a time estimation model.

In one embodiment, the computer program when executed by the processor implements machine learning training of the sample distribution information according to the sample training orders, and the step of obtaining the time estimation model is further configured to: acquiring a historical completion order different from the sample training order as an evaluation verification order; extracting historical distribution information from the evaluation verification order, wherein the historical distribution information comprises a historical warehouse address, a historical order address, historical waiting time and a historical vehicle state; inputting the historical warehouse address, the historical order address and the historical vehicle state into the time estimation model to obtain verification estimated time; and performing verification evaluation on the verification estimated time and the historical waiting time through evaluation standards, wherein the evaluation standards are at least one of the mean value of absolute deviation, the variance of absolute deviation, the median of absolute deviation and the R2 score.

In one embodiment, the computer program when executed by the processor for performing the step of obtaining the vehicle status and the warehouse address of the vendible vehicle is further configured to: when the number of the available vehicles is judged to be smaller than the number of the unmatched vehicle purchasing orders, acquiring the current time and the order placing time in the vehicle purchasing orders; ordering the vehicle purchasing orders according to the difference value between the current time and the ordering time; inputting the warehouse address, the order address and the vehicle state in the vehicle purchasing orders with the difference arranged in the front row into a trained time estimation model according to the quantity of the available vehicles, wherein the vehicle purchasing orders are arranged from large to small according to the difference.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. A vehicle deployment method, the method comprising:

receiving a vehicle purchasing order sent by a client terminal;

searching for a vehicle which can be sold and is matched with the vehicle requirement in a vehicle pool of which the first vehicle supply range corresponds to the order address, wherein the vehicle state of the vehicle which can be sold and sold is that a vehicle identification is associated with a license plate identification;

when the vehicle capable of being sold is matched, generating a vehicle distribution scheme according to a warehouse address of the vehicle capable of being sold and the order address of the vehicle purchase order;

acquiring the unmatched vehicle purchasing orders from an order pool according to a preset time period;

searching for available vehicles matched with the vehicle demands from a vehicle pool, and acquiring the vehicle states and warehouse addresses of the available vehicles;

inputting the warehouse address, the order address and the vehicle state into a trained time estimation model to obtain the waiting time of each saleable vehicle; the time estimation model completes order training according to history;

2. The method of claim 1, wherein said processing said warehouse address, said order address, and said wait time to arrive at a vehicle delivery schedule with a minimum total wait time comprises:

3. The method of claim 1, wherein said processing said warehouse address, said order address, and said wait time to arrive at a vehicle delivery schedule with a minimum total wait time comprises:

4. The method of claim 1, further comprising:

5. The method of claim 1, wherein the time estimation model is constructed in a manner that includes:

extracting sample distribution information from the sample training order;

6. The method of claim 5, wherein the machine learning training of the sample delivery information according to each sample training order to obtain a time estimation model further comprises:

7. The method of claim 1, further comprising, after said obtaining the vehicle status and the warehouse address of the vendable vehicle:

ordering the vehicle purchasing orders according to the difference value between the current time and the order placing time to obtain an order ordering result;

8. A vehicle blending apparatus, said apparatus comprising:

the vehicle allocation module is used for generating a vehicle distribution scheme according to the warehouse address of the marketable vehicle and the order address of the vehicle purchase order when the marketable vehicle is matched;

the vehicle purchasing order acquisition module is used for acquiring the unmatched vehicle purchasing orders from the order pool according to a preset time period;

the waiting time estimation module is used for inputting the warehouse address, the order address and the vehicle state in the vehicle purchasing order into a trained time estimation model to obtain the waiting time of each saleable vehicle; the time estimation model completes order training according to history;

9. The apparatus of claim 8, wherein the delivery plan generation module comprises:

10. The apparatus of claim 8, wherein the delivery plan generation module comprises:

11. The apparatus of claim 8, further comprising:

12. The apparatus of claim 8, wherein the latency pre-estimation module comprises:

13. The apparatus of claim 12, wherein the latency pre-estimation module comprises:

14. The apparatus of claim 8, further comprising:

15. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 7 when executing the computer program.

16. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.