WO2021129831A1 - 一种车辆调配方法、装置、设备及计算机可读存储介质 - Google Patents

一种车辆调配方法、装置、设备及计算机可读存储介质 Download PDF

Info

Publication number
WO2021129831A1
WO2021129831A1 PCT/CN2020/139589 CN2020139589W WO2021129831A1 WO 2021129831 A1 WO2021129831 A1 WO 2021129831A1 CN 2020139589 W CN2020139589 W CN 2020139589W WO 2021129831 A1 WO2021129831 A1 WO 2021129831A1
Authority
WO
WIPO (PCT)
Prior art keywords
vehicle
car rental
target car
target
model
Prior art date
Application number
PCT/CN2020/139589
Other languages
English (en)
French (fr)
Inventor
孟格思
李敏
吕伟
汪山人
薛淼
赵丛君
安康
王瑜
Original Assignee
北京嘀嘀无限科技发展有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from CN201911369388.1A external-priority patent/CN111815012A/zh
Priority claimed from CN202010006858.4A external-priority patent/CN111832872A/zh
Priority claimed from CN202010171414.6A external-priority patent/CN111383055B/zh
Application filed by 北京嘀嘀无限科技发展有限公司 filed Critical 北京嘀嘀无限科技发展有限公司
Publication of WO2021129831A1 publication Critical patent/WO2021129831A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/02—Reservations, e.g. for tickets, services or events
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising

Definitions

  • This application relates to the field of data processing technology, and in particular to a vehicle deployment method, device, equipment, and computer-readable storage medium.
  • the car rental platform In the car rental business scenario, the car rental platform generally determines the vehicle information such as the vehicle model according to the business needs, and then purchases the corresponding vehicle through the bidding method of the vehicle supplier such as the rental company, and then conducts the multi-channel inspection process on the purchased vehicle After inspection, it will be stored in the corresponding central warehouse of each target city. After the user purchases the vehicle that needs to be rented through the offline car rental store or online application, he completes the rental procedure according to the designated car viewing or pick-up location. Obviously, no matter which process is in the car rental business chain, it involves vehicle suppliers, central warehouses, car rental stores, rental companies, and other vehicle deployment issues based on vehicle business needs. When deploying vehicles, it is necessary to not only meet basic indicators such as the corresponding vehicle category and the number of requirements, but also to take into account the overall efficiency and even the detailed requirements of vehicle deployment among multiple parties.
  • the vehicle allocation method includes: acquiring associated feature data of a target car rental point, the associated feature data including related feature data of at least one of the target car rental point and a target car rental area; the target car rental area is related to the The area related to the target car rental location; processing the associated feature data of the target car rental location based on a predictive model to determine the associated vehicle demand of the target car rental location, where the associated vehicle demand is the target car rental location at the target car rental The relevant demand for vehicles in the area; at least based on the associated vehicle demand, a vehicle scheduling plan for the target car rental point is determined.
  • a vehicle deployment device including a processor
  • the processor includes: an acquisition module, configured to: acquire associated feature data of a target car rental point, the associated feature data including the target car rental point and Related feature data of at least one of the target car rental areas; the target car rental area is an area related to the target car rental location; the first determining module is configured to: associate feature data of the target car rental location based on a prediction model Processing is performed to determine the associated vehicle demand of the target car rental point, where the associated vehicle demand is the relevant demand of the target car rental point for vehicles in the target car rental area; the second determining module is configured to: at least based on the Associate the vehicle demand, and determine the vehicle scheduling plan of the target car rental point.
  • One of the embodiments of the present application provides a vehicle deployment device, including: a memory; a processor; and a computer program; wherein the computer program is stored in the memory and is configured to be executed by the processor to achieve The method described above.
  • One of the embodiments of the present application provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the method as described above is implemented.
  • FIG. 1 is a schematic diagram of an application scenario of a vehicle deployment system 100 according to some embodiments of the present application;
  • FIG. 2 is a schematic diagram of exemplary hardware and/or software of an exemplary computing device 200 according to some embodiments of the present application;
  • FIG. 3 is a schematic diagram of exemplary hardware and/or software of an exemplary mobile device 300 according to some embodiments of the present application;
  • FIG. 4 is an exemplary flowchart of a vehicle deployment method 400 according to some embodiments of the present application.
  • FIG. 5 is an exemplary structure diagram of a vehicle deployment device 500 according to some embodiments of the present application.
  • FIG. 6 is an exemplary structure diagram of a vehicle deployment device 600 according to some embodiments of the present application.
  • FIG. 7 is an exemplary flowchart of a vehicle deployment method 700 according to some embodiments of the present application.
  • FIG. 8 is an exemplary flowchart of a vehicle deployment method 800 according to some embodiments of the present application.
  • FIG. 9 is an exemplary flowchart of a vehicle deployment method 900 according to some embodiments of the present application.
  • FIG. 10 is an exemplary flowchart of a vehicle deployment method 1000 according to some embodiments of the present application.
  • FIG. 11 is an exemplary flowchart of a vehicle deployment method 1100 according to some embodiments of the present application.
  • FIG. 12 is an exemplary structure diagram of a vehicle deployment device 1200 according to some embodiments of the present application.
  • FIG. 13 is an exemplary structure diagram of an electronic device 1300 for vehicle deployment according to some embodiments of the present application.
  • FIG. 14 is an exemplary flowchart of a vehicle deployment method 1400 according to some embodiments of the present application.
  • FIG. 15 is an exemplary flowchart of a vehicle deployment method 1500 according to some embodiments of the present application.
  • FIG. 16 is an exemplary flowchart of a vehicle deployment method 1600 according to some embodiments of the present application.
  • FIG. 17 is a schematic diagram of an application scenario in which the vehicle deployment method and its hardware device according to some embodiments of the present application are applied to a user to use a car rental platform;
  • FIG. 18 is an exemplary flowchart of a vehicle deployment method 1800 according to some embodiments of the present application.
  • FIG. 19 is an exemplary flowchart of a vehicle deployment method 1900 according to some embodiments of the present application.
  • FIG. 20 is a schematic flowchart of a vehicle deployment method 2000 according to some embodiments of the present application.
  • FIG. 21 is an exemplary flowchart of a vehicle deployment method 2100 according to some embodiments of the present application.
  • FIG. 22 is an example diagram of an application scenario of a vehicle deployment method 2100 according to some embodiments of the present application.
  • FIG. 23 is an exemplary structure diagram of a vehicle deployment device 2300 according to some embodiments of the present application.
  • Fig. 24 is an exemplary structure diagram of a vehicle deployment device 2400 according to some embodiments of the present application.
  • FIG. 25 is an exemplary flowchart of a vehicle deployment method 2500 according to some embodiments of the present application.
  • FIG. 26 is an exemplary flowchart of a vehicle deployment method 2600 according to some embodiments of the present application.
  • FIG. 27 is an exemplary flowchart of a vehicle deployment method 2700 according to some embodiments of the present application.
  • FIG. 28 is an exemplary flowchart of a vehicle deployment method 2800 according to some embodiments of the present application.
  • FIG. 29 is an exemplary flowchart of a vehicle deployment method 2900 according to some embodiments of the present application.
  • FIG. 30 is a schematic diagram of an exemplary system implemented by a vehicle deployment method according to some embodiments of the present application.
  • FIG. 31 is an exemplary structure diagram of a vehicle deployment device 3100 provided according to some embodiments of the present application.
  • FIG. 32 is an exemplary flowchart of a prediction model training process 3200 according to some embodiments of the present application.
  • system is a method for distinguishing different components, elements, parts, parts, or assemblies of different levels.
  • the words can be replaced by other expressions.
  • FIG. 1 is a schematic diagram of an application scenario of a vehicle deployment system 100 according to some embodiments of the present application.
  • the application scenario may include the server 110, the network 120, the data query terminal 130, the database 140, and other data sources 150.
  • the server 110 may include a processing device 112.
  • the vehicle deployment system 100 may perform vehicle deployment by implementing the methods and/or processes disclosed in this application.
  • the server 110 can be deployed on a public car rental platform, or in a target car rental area or a target car rental point.
  • the server 110 can be used to obtain the association of the target car rental point from the database 140 or other data sources 150 via the network 120. Characteristic data, and perform corresponding data processing to obtain vehicle scheduling plan data for the target car rental point.
  • the data in the database 140 and other data sources 150 can come from a car rental platform that uploads data related to the vehicle deployment business generated by the target car rental location and the target car rental area, or it can come from the target car rental location or the target car rental area.
  • the data content includes the associated feature data of the target car rental location, associated demand data, vehicle scheduling plan data and other vehicle deployment business related data, vehicle deployment business related data in the target car rental area, prediction model training sample data, prediction model result data, and so on.
  • the vehicle deployment system 100 can be used for vehicle deployment associated feature data based on target offline car rental stores and target central warehouses, vehicle deployment associated characteristic data of designated areas between the target central warehouse and target offline car rental stores, and target
  • the leasing company and its designated storage vehicle target central warehouse vehicle deployment associated feature data, or the target central warehouse and target city vehicle deployment associated feature data and other vehicle deployment scenarios require data processing, so as to obtain the corresponding vehicle scheduling plan.
  • the server 110 can obtain the associated feature data of the target leasing company from the database 140 and other data sources 150 (For example, it can be the historical model data of the target leasing company, historical location demand data, etc.), and then through the execution of the program instructions to perform the data processing process by the predictive model, and finally determine the target leasing company's vehicle scheduling plan to guide the target leasing company How to assign vehicles to the target central warehouse where the vehicles are stored.
  • the database 140 and other data sources 150 can be the historical model data of the target leasing company, historical location demand data, etc.
  • All business-related data generated during the vehicle deployment business can be stored in the server 110 and/or the database 140 and other data sources 150.
  • the prediction model and its corresponding data can be directly stored in the server 110 or other storage devices (set according to the needs of the scene), or can be stored in the database 140 or other data sources 150.
  • the users of the data query terminal 130 may include central warehouses, offline car rental stores, rental companies, car rental platforms, users who need to rent a car, or other possible users related to car rental business or vehicle deployment.
  • the data query authority it can be carried out according to specific application scenarios. Set accordingly.
  • the server 110 and the data query terminal 130 may be connected through the network 120, and the database 140 may be connected with the server 110 through the network 120, or may be directly connected to the server 110 or located inside the server 110.
  • the database 140 and other data sources 150 can be connected to the network 120 to communicate with one or more components of the vehicle deployment system 100.
  • One or more components of the vehicle deployment system 100 can access data or instructions stored in the data query terminal 130, the database 140, and other data sources 150 through the network 120.
  • the server 110, the data query terminal 130, and other possible system components may include a database 140.
  • the server 110, the data query terminal 130, and other possible system components may include a processor.
  • the server 110 may be used to manage resources and process data and/or information from at least one component of the system or an external data source (for example, a cloud data center).
  • the server 110 may be a single server or a group of servers.
  • the server group may be centralized or distributed (for example, the server 110 may be a distributed system), it may be dedicated, or other devices or systems may provide services at the same time.
  • the server 110 may be regional or remote.
  • the server 110 may be implemented on a cloud platform or provided in a virtual manner.
  • the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-layer cloud, etc., or any combination thereof.
  • the server 110 may include a processing device 112.
  • the processing device 112 may adopt a corresponding processor to process data and/or information obtained from other devices or system components.
  • the processor may execute program instructions based on these data, information, and/or processing results to perform one or more functions described in this application.
  • the processing device 112 may include one or more sub-processing devices (for example, a single-core processing device or a multi-core and multi-core processing device).
  • the processing device 112 may include a central processing unit (CPU), an application specific integrated circuit (ASIC), an application specific instruction processor (ASIP), a graphics processing unit (GPU), a physical processor (PPU), a digital signal processor ( DSP), Field Programmable Gate Array (FPGA), Editable Logic Circuit (PLD), Controller, Microcontroller Unit, Reduced Instruction Set Computer (RISC), Microprocessor, etc. or any combination of the above.
  • CPU central processing unit
  • ASIC application specific integrated circuit
  • ASIP application specific instruction processor
  • GPU graphics processing unit
  • PPU physical processor
  • DSP digital signal processor
  • FPGA Field Programmable Gate Array
  • PLD Editable Logic Circuit
  • Controller Microcontroller Unit
  • RISC Reduced Instruction Set Computer
  • the network 120 may connect various components of the system and/or connect the system and external resource parts.
  • the network 120 enables communication between various components and with other parts outside the system, facilitating the exchange of data and/or information.
  • the network 120 may be any one or more of a wired network or a wireless network.
  • the network 120 may include a cable network, a fiber optic network, a telecommunication network, the Internet, a local area network (LAN), a wide area network (WAN), a wireless local area network (WLAN), a metropolitan area network (MAN), and a public switched telephone network (PSTN) , Bluetooth network, ZigBee network (ZigBee), near field communication (NFC), intra-device bus, intra-device wiring, cable connection, etc.
  • LAN local area network
  • WAN wide area network
  • WLAN wireless local area network
  • MAN metropolitan area network
  • PSTN public switched telephone network
  • Bluetooth network ZigBee network
  • NFC near field communication
  • intra-device bus intra-device wiring, cable
  • the network connection between the various parts can be in one of the above-mentioned ways, or in multiple ways.
  • the network may be a variety of topological structures such as point-to-point, shared, and centralized, or a combination of multiple topologies.
  • the network 120 may include one or more network access points.
  • the network 120 may include wired or wireless network access points, such as base stations and/or network exchange points 120-1, 120-2, ..., through which one or more components of the system 100 can be connected to the network 120 To exchange data and/or information.
  • the data query terminal 130 refers to one or more terminal devices or software used for data query.
  • the data query terminal 130 may be used by one or more users, may include users who directly use the service, or may include other related users.
  • the data query terminal 130 may be one of the mobile device 130-1, tablet computer 130-2, laptop computer 130-3, desktop computer and other devices with input and/or output functions, or Any combination of it.
  • the mobile device 130-1 may include a wearable device, a smart mobile device, etc., or any combination thereof.
  • the smart mobile device may include a smart phone, a personal digital assistant (PDA), a game device, a navigation device, a handheld terminal (POS), etc., or any combination thereof.
  • PDA personal digital assistant
  • POS handheld terminal
  • the desktop computer 130-4 may be a vehicle-mounted computer, a vehicle-mounted TV, or the like.
  • the other device mobile device 130-1 with input and/or output functions may include a dedicated question and answer terminal set in a public place.
  • the database 140 may be used to store data and/or instructions.
  • the database 140 is implemented in a single central server, multiple servers connected through a communication link, or multiple personal devices.
  • the database 140 may include mass memory, removable memory, volatile read-write memory (for example, random access memory RAM), read-only memory (ROM), etc., or any combination thereof.
  • the mass storage device may include magnetic disks, optical disks, solid-state disks, and the like.
  • the database 140 may be implemented on a cloud platform.
  • Other data sources 150 may be used to provide one or more sources of other information to the system.
  • the other data source 150 can be one or more devices, can be one or more application program interfaces, can be one or more database query interfaces, can be one or more protocol-based information acquisition interfaces, and can be other accessible
  • the information method can be a combination of two or more of the above methods.
  • the information provided by the information source may already exist when the information is extracted, or it may be generated temporarily when the information is extracted, or a combination of the above methods.
  • other data sources 150 may be used to provide the system with environmental information, weather information, or any other possible information related to car rental business or vehicle deployment.
  • FIG. 2 is a schematic diagram of exemplary hardware and/or software of an exemplary computing device 200 according to some embodiments of the present application.
  • the server 110 or the data query terminal 130 may be implemented on the computing device 200.
  • the processing device 112 may implement and execute the functions of the processing device 112 disclosed in this specification on the computing device 200.
  • the computing device 200 may include a bus 210, a processor 220, a read-only memory 230, a random access memory 240, a communication port 250, an input/output 260, and a hard disk 270.
  • the processor 220 can execute calculation instructions (program codes) and perform the functions of the vehicle deployment system 100 described in this specification.
  • the calculation instructions may include programs, objects, components, data structures, procedures, modules, functions (the functions refer to the specific functions described in this specification), and the like.
  • the processor 220 may process image or text data obtained from any other components of the vehicle deployment system 100.
  • the processor 220 may include a microcontroller, a microprocessor, a reduced instruction set computer (RISC), an application specific integrated circuit (ASIC), an application specific instruction set processor (ASIP), a central processing unit (CPU) , Graphics processing unit (GPU), physical processing unit (PPU), microcontroller unit, digital signal processor (DSP), field programmable gate array (FPGA), advanced RISC machine (ARM), programmable logic device, and Any circuits and processors that perform one or more functions, etc., or any combination thereof.
  • the computing device 200 in FIG. 2 only describes one processor, but it should be noted that the computing device 200 in this specification may also include multiple processors.
  • the memory of the computing device 200 may store data/information acquired from any other components of the vehicle deployment system 100.
  • exemplary ROMs may include mask ROM (MROM), programmable ROM (PROM), erasable programmable ROM (PEROM), electrically erasable programmable ROM (EEPROM), compact disk ROM (CD-ROM), and digital Universal disk ROM, etc.
  • Exemplary RAM may include dynamic RAM (DRAM), double rate synchronous dynamic RAM (DDR SDRAM), static RAM (SRAM), thyristor RAM (T-RAM), zero capacitance (Z-RAM), and the like.
  • the input/output 260 may be used to input or output signals, data or information.
  • the input/output 260 may include an input device and an output device.
  • Exemplary input devices may include a keyboard, a mouse, a touch screen, a microphone, etc., or any combination thereof.
  • Exemplary output devices may include display devices, speakers, printers, projectors, etc., or any combination thereof.
  • Exemplary display devices may include liquid crystal displays (LCD), light emitting diode (LED) based displays, flat panel displays, curved displays, television equipment, cathode ray tubes (CRT), etc., or any combination thereof.
  • LCD liquid crystal displays
  • LED light emitting diode
  • CRT cathode ray tubes
  • the communication port 250 can be connected to a network for data communication.
  • the connection can be a wired connection, a wireless connection, or a combination of both.
  • Wired connections can include cables, optical cables, telephone lines, etc., or any combination thereof.
  • the wireless connection may include Bluetooth, Wi-Fi, WiMax, WLAN, ZigBee, mobile networks (for example, 3G, 4G, or 5G, etc.), etc., or any combination thereof.
  • the communication port 250 may be a standardized port, such as RS232, RS485, and so on. In some embodiments, the communication port 250 may be a specially designed port.
  • FIG. 3 is a schematic diagram of exemplary hardware and/or software of an exemplary mobile device 300 according to some embodiments of the present application.
  • the mobile device 300 may include a communication unit 310, a display unit 320, a graphics processing unit (GPU) 330, a processor (CPU) 340, an input/output unit 350, a memory 360, a storage unit 370, and the like.
  • the mobile device 300 may also include any other suitable components, including but not limited to a system bus or a controller (not shown in the figure).
  • the operating system 361 for example, iOS, Android, Windows Phone, etc.
  • the application program 362 may be loaded from the storage unit 370 into the memory 360 so as to be executed by the CPU 340.
  • the application program 362 may include a browser or an application program for receiving text, images, audio, or other related information from the vehicle deployment system 100. The user interaction of the information flow may be implemented through the input/output unit 350 and provided to the processing device 112 and/or other components of the vehicle deployment system 100 through the network 120.
  • a computing device or a mobile device can be used as a hardware platform for one or more components described in this specification.
  • the hardware components, operating systems, and programming languages of these computers or mobile devices are conventional in nature, and those skilled in the art can adapt these technologies to the system described in this specification after being familiar with these technologies.
  • a computer with user interface components can be used to implement a personal computer (PC) or other types of workstations or terminal devices. If properly programmed, the computer can also act as a server.
  • FIG. 4 is an exemplary flowchart of a vehicle deployment method 400 according to some embodiments of the present application. As shown in FIG. 4, the method 400 includes the following steps. In some embodiments, the method 400 may be executed by a processor.
  • Step 410 Obtain associated feature data of the target car rental location, where the associated feature data includes related feature data of at least one of the target car rental location and the target car rental area; the target car rental area is an area related to the target car rental location.
  • the target car rental point can be a car rental point or a vehicle storage location where you can check and/or pick up the car offline.
  • it can be an offline car rental store, a central warehouse area, or a specific central warehouse, or it can be a place that supplies or distributes vehicles. Leasing company.
  • an offline car rental store may be a place that provides customers with vehicles and after-sales service.
  • the central bin area may be a delimited area containing one or more central bins.
  • the central warehouse may be a place for vehicle entry and exit detection, parking, scheduling, and pickup, including multiple storage locations, which are used to store vehicles.
  • the rental company may be a merchant that provides vehicles for the rental car platform.
  • the target car rental area may be an area related to the target car rental location.
  • the target car rental area can be a car rental area related to the geographic location of the target car rental point or business connection, for example, it can be an offline car rental store, a central warehouse area, or a radially demarcated area around a specific central warehouse, or they can be mutually exclusive
  • the target car rental area can be a central warehouse area or a central warehouse that is geographically close to an offline car rental store or business-related
  • the target car rental area can be a central warehouse area or a central warehouse that is geographically nearby or business-related to the central warehouse.
  • the target car rental area may also be a central warehouse or a central warehouse area that is associated with the rental company's business (for example, has an association relationship for specifying the storage of the rental company's vehicles, etc.) or geographic location.
  • the target car rental area may be an offline car rental store that is associated with the rental company's business or geographic location.
  • the target car rental area may be an offline car rental store that is geographically close to the rental company.
  • other combinations of the target car rental location and the target car rental area can be set according to actual needs, that is, offline car rental stores, central warehouses, central warehouse areas, rental companies, target cities, and the geographic location of any one of the above Location-related or business-related car rental areas (the car rental area can be an artificially demarcated area according to business needs, such as a similar surrounding radiation delineated area), select any two of them and set them as the target car rental location and the target car rental area.
  • the target car rental area may be a location area that includes a target car rental point, such as the location area of an offline car rental store, that is, an area near the target car rental point, and the target car rental area is determined according to the target car rental point.
  • a target car rental point such as the location area of an offline car rental store, that is, an area near the target car rental point, and the target car rental area is determined according to the target car rental point.
  • the target car rental area may be a target city including at least one central warehouse.
  • the characteristic data corresponding to the target city has the characteristics of the city or each selected central warehouse.
  • the target car rental area may be a central warehouse area or a central warehouse corresponding to the rental company for vehicle storage.
  • the rental company generally has multiple central warehouses. According to the rental company’s car rental business conditions (including vehicle richness, brand Influence, hot models, etc.) determine the central warehouse area or central warehouse requirements for storing vehicles.
  • car rental business conditions including vehicle richness, brand Influence, hot models, etc.
  • the associated feature data may be related feature data of at least one of the target car rental point and the target car rental area.
  • the associated feature data may be any possible type of information data that can affect the business requirements, vehicle requirements, and/or vehicle deployment requirements of the target car rental location or the target car rental area in each link of the car rental.
  • the associated feature data may be the first historical vehicle attribute information of the rented vehicle at the target car rental location and the second historical vehicle attribute information corresponding to the historical car rental order in the target car rental area.
  • the first historical vehicle attribute information may include each rented vehicle.
  • the second historical vehicle attribute information may include the model information of the rental vehicle included in each historical car rental order; the associated feature data may also be the environmental information of the target car rental location and the selection-related feature data of the target car rental area.
  • the product selection-related feature data of the rental car area can include multiple dimensions of historical product selection demand-related data in the target car rental area; the associated feature data can also be the historical turnover vehicle information and related information of the target car rental location, and the location information of the target car rental area.
  • Historical turnover vehicle information and related information may include the number of historical turnover vehicles and multi-dimensional vehicle turnover impact data, and the location information may include the number of location gaps.
  • the first historical vehicle attribute information and the second historical vehicle attribute information refer to vehicle corresponding attribute information and the number of vehicles that can indicate the model, brand, or other vehicle indicator of a specific vehicle.
  • the first historical vehicle attribute information may include historical model information corresponding to each rented vehicle in the target car rental location
  • the second historical vehicle attribute information may include the model of the rental vehicle included in each historical car rental order in the target car rental area.
  • Information For more examples of historical model information corresponding to each rented vehicle at the target car rental point, and each historical car rental order in the target car rental area, for more examples of the model information of the rented vehicle included, please refer to Figure 11 and the corresponding description, which will not be repeated here. Go into details.
  • Environmental information refers to data on the dimensions of objective factors that can affect the demand for car models at the target car rental point.
  • the environmental information may reflect environmental information such as the geographic location and scale of the target car rental point covering the car rental area.
  • the environmental information please refer to FIG. 18 and corresponding descriptions, which will not be repeated here.
  • Product selection-related feature data refers to corresponding feature data that can reflect the selection of vehicles in the target car rental area.
  • Product selection refers to determining that each target car rental point corresponds to multiple pre-selected models and the demand for each pre-selected model.
  • the product selection-related feature data may include multiple dimensions of historical product selection demand-related data in the target car rental area, and the multiple dimensions of historical product selection demand-related data in the target car rental area may be feature data corresponding to the target city.
  • feature data corresponding to the target city refer to FIG. 18 and the corresponding description, which will not be repeated here.
  • Historical turnover vehicle information and related information refer to data that can reflect the historical turnover of the target car rental location and other factors that can affect the situation of turnover vehicles.
  • historical turnover information and related information can include the number of historical turnover vehicles and the impact of multi-dimensional vehicle turnover. data.
  • historical turnover vehicle information and related information may include the information dimension data and time data of the leasing company, and further may include the total number of rental cars and the total number of cars returned by the leasing company, the number of historical rental car orders, and the scale data of the leasing company. Dimensional features such as information on holidays, holidays, weather, rental company credit scores, model richness, and rental brand awareness.
  • historical turnover vehicle information and related information please refer to Figure 29, Figure 30 and their corresponding descriptions. Go into details again.
  • the location information refers to the location planning or setting situation for storing vehicles.
  • the location information of the target car rental area may include the number of gaps in the location of the target car rental area.
  • the location information of the target car rental area may be the number of gapped vehicles in each central warehouse of the leasing company.
  • FIGS. 28 to 30 please refer to FIGS. 28 to 30 and The corresponding description will not be repeated here.
  • the associated feature data may be the first historical vehicle attribute information of the rented vehicle at the target car rental point and the selection-related feature data of the target car rental area.
  • the associated feature data may be the history corresponding to each rented vehicle. Model information and data related to historical product selection requirements in multiple dimensions in the target car rental area.
  • the associated feature data may be the first historical vehicle attribute information of the rented vehicle at the target car rental location and the location information of the target car rental area.
  • the associated feature data may be historical model information corresponding to each rented vehicle. And the number of location gaps.
  • the associated feature data may be the historical turnover vehicle information of the target car rental location and related information and the second historical vehicle attribute information corresponding to the historical car rental orders in the target car rental area.
  • the associated feature data may be the number of historical turnover vehicles And multi-dimensional vehicle turnover impact data and the model information of the leased vehicles included in each historical car rental order.
  • the associated feature data can select any two of the following groups for combination settings: the first historical vehicle attribute information of the rented vehicle at the target car rental point, the environmental information of the target car rental point, and the historical turnover of the target car rental point The vehicle information and related information, the second historical vehicle attribute information corresponding to the historical car rental order in the target car rental area, the selection-related feature data of the target car rental area, and the storage location information of the target car rental area.
  • the processor can obtain the associated feature data of the target car rental point in a variety of ways.
  • the historical vehicle attribute information in the associated feature data can be obtained by uploading the target car rental point to the server.
  • the associated feature data of the corresponding target car rental point can also be obtained through the server of the car rental platform.
  • Step 420 Process the associated feature data of the target car rental spot based on the predictive model, and determine the associated vehicle demand of the target car rental spot.
  • the associated car demand is the relevant demand of the target car rental spot for vehicles in the target car rental area.
  • the predictive model is a specific calculation model or algorithm used to obtain the corresponding associated vehicle demand through the calculation and processing of the associated feature data of the target car rental point.
  • the prediction model can use artificial intelligence algorithms, specifically, decision trees, random forests, logistic regression, support vector machines, naive Bayes, K-nearest neighbor algorithm, K-means algorithm, Adaboost (a kind of Boosting algorithm), Neural network, Markov's machine learning algorithm.
  • the prediction model may be created using a decision tree algorithm. For more embodiments of building a prediction model through the decision tree algorithm, refer to FIG. 19 and the corresponding description, which will not be repeated here.
  • step 420 may be implemented as the following process:
  • the first historical vehicle attribute information of the rented vehicle at the target car rental location and the second historical vehicle attribute information corresponding to the historical car rental order in the target car rental area are processed to determine the model demand of the target car rental location.
  • the associated feature data includes: the first historical vehicle attribute information of the rented vehicle at the target car rental point and the second historical vehicle attribute information corresponding to the historical car rental order in the target car rental area; the associated vehicle demand of the target car rental point includes the information of the target car rental point Model demand.
  • step 420 may be implemented as the following process:
  • the environmental information of the target car rental location and the product-related feature data of the target car rental area are processed to determine the model demand of the target car rental location.
  • the associated feature data includes: environmental information of the target car rental location and product-related feature data of the target car rental area; the associated vehicle demand of the target car rental site includes the model demand of the target car rental site.
  • step 420 may be implemented as the following process:
  • the historical turnover vehicle information and related information of the target car rental location, and the location information of the target car rental area are processed, and the number of vehicles required for the target car rental location is determined.
  • the associated feature data includes: historical turnover vehicle information and related information of the target car rental point, and location information of the target car rental area; the associated vehicle demand of the target car rental point includes the number of vehicles demand of the target car rental point.
  • Step 430 Determine a vehicle scheduling plan for the target car rental point based at least on the associated vehicle demand.
  • the vehicle scheduling plan refers to the vehicle scheduling related information data that can be used to guide the vehicle deployment operation for the target car rental point.
  • the vehicle scheduling plan may include vehicle information to be deployed, the vehicle information to be deployed may include attribute information of the vehicle to be deployed and the number of vehicles of each attribute, and the vehicle attribute information may include model, brand or any other possible vehicle attribute information.
  • the vehicle scheduling plan of the target car rental location in addition to determining the vehicle scheduling plan of the target car rental location based only on the associated vehicle demand, it can also comprehensively consider the statistics of the vehicle callback situation data in the vehicle deployment business, the target car rental location or the user's special vehicle demand data, and the target car rental Regional environmental change factor data or any other corresponding dimensional data that may affect vehicle deployment in the car rental business.
  • the above-mentioned step 430 can be implemented as the following process:
  • the dispatched vehicles of the vehicle dispatch plan are determined from the vehicles to be dispatched.
  • the third vehicle attribute information refers to vehicle corresponding attribute information that can indicate the model, brand, or other vehicle indicator of the vehicle to be dispatched.
  • the third vehicle attribute information may be the model of the vehicle to be dispatched and the number of vehicles.
  • the dispatcher can select the target model number matching the target model number of the central warehouse/offline store vehicle.
  • the above-mentioned step 430 can be implemented as the following process:
  • the dispatched vehicles of the vehicle dispatch plan are determined from the vehicles to be dispatched. For example, based on the third vehicle attribute information of the vehicles to be dispatched and the number of vehicles quota demand of the leasing company, the dispatched vehicles of the vehicle dispatch plan may be determined from the vehicles to be dispatched.
  • the vehicle deployment method, device, equipment, and computer-readable storage medium provided by the embodiments of the present application acquire the associated feature data of the target car rental site that affects the vehicle deployment business, use the predictive model to process and determine the associated vehicle demand of the target car rental site, and then according to The associated vehicle needs determine its vehicle scheduling plan, providing an efficient vehicle allocation plan that is more reasonable and meets the needs of multiple car rental business scenarios to save business operating costs, meet multiple needs of users, and improve user experience.
  • FIG. 5 is an exemplary structural diagram of a vehicle deployment device 500 according to some embodiments of the present application.
  • the vehicle deployment device 500 includes an acquisition module 501, a first determination module 502, and a second determination module 503.
  • the obtaining module 501 is configured to: obtain associated feature data of a target car rental location, the associated feature data including related feature data of at least one of the target car rental location and the target car rental area; the target car rental area is an area related to the target car rental location.
  • the obtaining module 501 may be used to obtain the location information of the target car rental point and the historical vehicle attribute information of the rented vehicle at the target car rental point, determine the location area corresponding to the target car rental point, and obtain the positioning The information belongs to the historical car rental order corresponding to the user terminal in the location area.
  • the acquiring module 501 may be used to acquire characteristic data of a target city in a first preset historical time period, where the target city is a city with at least one central warehouse, and the characteristic data is used for Represents the environmental information of each central warehouse in the target city and the associated data of the vehicle type in the target city.
  • the obtaining module 501 may be used to obtain data from various parties, and perform data cleaning, feature processing, and structured storage.
  • the first determining module 502 is configured to process the associated feature data of the target car rental location based on the predictive model, and determine the associated vehicle demand of the target car rental point.
  • the associated vehicle demand is the relevant demand of the target car rental point for vehicles in the target car rental area. For more details about determining the associated vehicle demand of the target car rental location based on the prediction model, refer to FIG. 4 and related descriptions, which will not be repeated here.
  • the first determining module 502 may be used to predict the demand for car rental and the number of cars returned through the prediction model to provide decision-making data support for subsequent modules.
  • the second determining module 503 is configured to determine a vehicle scheduling plan of the target car rental point based at least on the demand of the associated vehicle. For more details about the vehicle scheduling scheme for determining the target car rental point, refer to Fig. 4 and related descriptions, which will not be repeated here. It can be used for its forecast-based data and business scenarios to make automated real-time decisions.
  • the second determining module 503 may be used to make automated real-time decisions based on predicted data and business scenarios.
  • the above description of the vehicle deployment device 500 and its modules is only for convenience of description, and does not limit the present application within the scope of the embodiments mentioned. It can be understood that for those skilled in the art, after understanding the principle of the system, it is possible to arbitrarily combine various modules, or form a subsystem to connect with other modules without departing from this principle.
  • the acquiring module 501, the first determining module 502, and the second determining module 503 disclosed in FIG. 5 may be different modules in a system, or may be one module to implement the above two or more modules. Function. For example, each module may share a storage module, and each module may also have its own storage module. Such deformations are all within the protection scope of this application.
  • FIG. 6 is an exemplary structure diagram of a vehicle deployment device 600 according to some embodiments of the present application.
  • some embodiments of the present application also provide a vehicle deployment device 600.
  • the vehicle deployment device 600 includes a memory 601, a processor 602, and a computer program.
  • the computer program is stored in the memory 601 and configured to be used by the processor.
  • 602 executes the vehicle deployment method provided in any of the foregoing embodiments. For specific details of the executed vehicle deployment method process, refer to the foregoing method embodiment, which will not be repeated here.
  • the processor may be a central processing unit (English: Central Processing Unit, abbreviated as: CPU), or other general-purpose processors, digital signal processors (English: Digital Signal Processor, referred to as DSP), application specific integrated circuit (English: Application Specific Integrated Circuit, referred to as ASIC), etc.
  • the general-purpose processor may be a microprocessor or the processor may also be any conventional processor or the like. The steps of the method disclosed in combination with the invention can be directly embodied as executed and completed by a hardware processor, or executed by a combination of hardware and software modules in the processor.
  • the memory may include a high-speed RAM memory, and may also include a non-volatile storage NVM, such as at least one disk memory.
  • NVM non-volatile storage
  • the bus can be an Industry Standard Architecture (ISA) bus, Peripheral Component Interconnect (PCI) bus, or Extended Industry Standard Architecture (EISA) bus, etc.
  • ISA Industry Standard Architecture
  • PCI Peripheral Component Interconnect
  • EISA Extended Industry Standard Architecture
  • the bus can be divided into address bus, data bus, control bus and so on.
  • the buses in the drawings of this application are not limited to only one bus or one type of bus.
  • an embodiment of the present application also provides a computer-readable storage medium on which a computer program is stored, and the computer program is executed by a processor to implement the vehicle deployment method described in the foregoing embodiment.
  • the above-mentioned computer-readable storage medium may be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable and removable Programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.
  • SRAM static random access memory
  • EEPROM electrically erasable and removable Programmable read-only memory
  • EPROM erasable programmable read-only memory
  • PROM programmable read-only memory
  • ROM read-only memory
  • magnetic memory flash memory
  • flash memory magnetic disk or optical disk.
  • a readable storage medium may be any available medium that can be accessed by a general purpose or special purpose computer.
  • An exemplary readable storage medium is coupled to the processor, so that the processor can read information from the readable storage medium and can write information to the readable storage medium.
  • the readable storage medium may also be an integral part of the processor.
  • the processor and the readable storage medium may be located in Application Specific Integrated Circuits (ASIC for short).
  • ASIC Application Specific Integrated Circuits
  • the processor and the readable storage medium may also exist in the device as discrete components.
  • FIG. 7 is an exemplary flowchart of a vehicle deployment method 700 according to some embodiments of the present application.
  • the method 700 may be executed by a processor.
  • Step 710 Obtain associated feature data of the target car rental location, where the associated feature data includes: the first historical vehicle attribute information of the rented vehicle at the target car rental point and the second historical vehicle attribute information corresponding to the historical car rental order in the target car rental area.
  • Step 720 Process the first historical vehicle attribute information of the rented vehicle at the target car rental location and the second historical vehicle attribute information corresponding to the historical car rental order in the target car rental area based on the predictive model, and determine the model demand of the target car rental point.
  • Step 730 Determine a vehicle scheduling plan for the target car rental spot based on at least the vehicle type demand of the target car rental spot.
  • the vehicle scheduling method provided in this embodiment obtains the associated feature data of the target car rental point, including: the first historical vehicle attribute information of the rented vehicle and the second historical vehicle attribute information corresponding to the historical car rental order in the target car rental area, based on predictions
  • the model processes the associated feature data, confirms the model requirements of the target car rental point, and further confirms the scheduling plan of the target car rental point, which can refine the car model demand of the target car rental point, improve the matching degree of vehicle scheduling with the actual model demand, and improve the vehicle
  • the deployment efficiency is conducive to the improvement of car rental efficiency at car rental locations.
  • FIG. 8 is an exemplary flowchart of a vehicle deployment method 800 according to some embodiments of the present application.
  • the method 800 may be executed by a processor.
  • Step 810 Obtain associated feature data of the target car rental location, where the associated feature data includes: first historical vehicle attribute information of the rented vehicle at the target car rental point and second historical vehicle attribute information corresponding to the historical car rental order in the target car rental area;
  • Step 820 Process the first historical vehicle attribute information of the rented vehicle at the target car rental location and the second historical vehicle attribute information corresponding to the historical car rental order in the target car rental area based on the prediction model, and determine the model demand of the target car rental location;
  • step 830 based on the third vehicle attribute information of the vehicles to be dispatched and the model requirements of the target car rental point, the dispatched vehicles of the vehicle dispatch plan are determined from the vehicles to be dispatched.
  • FIG. 9 is an exemplary flowchart of a vehicle deployment method 900 according to some embodiments of the present application.
  • the method 900 may be executed by a processor.
  • Step 910 Obtain the associated feature data of the target car rental location.
  • the associated feature data includes: historical model information corresponding to each leased vehicle of the leased vehicle at the target car rental location and the rental included in the historical car rental order corresponding to the historical car rental order in the target car rental area Vehicle model information;
  • Step 920 Based on the predictive model, process the historical model information corresponding to each rented vehicle at the target car rental site and the model information of the rented vehicle included in each historical car rental order, and determine the target car model at the target car rental site and the number of target car models corresponding to it;
  • Step 930 Based on the third vehicle attribute information of the vehicles to be dispatched and the target car model of the target car rental site and the number of corresponding target car models, select a dispatched vehicle matching the target car model of the target car rental site and the number of target car models.
  • the vehicle scheduling method provided in this embodiment obtains associated feature data including historical model information corresponding to each leased vehicle and model information of the leased vehicle included in each historical car rental order, and processes the associated feature data based on a predictive model. Determine the target car model of the target car rental point that is its corresponding number, and determine the dispatch vehicle based on this as the target car rental point.
  • This method allows users to watch and pick up the car directly at the car rental point without going to the target car rental area, meets the user's demand for seeing and picking up the car nearby, saves the user's time, and improves the user's car rental experience. The user brings the car to the rental car spot to see and pick up the car, which also increases the visit rate of the car rental spot and improves the car rental efficiency of the car rental spot.
  • the associated feature data of the target car rental point may select the historical vehicle attribute information of the rented vehicle at the target car rental point and the rental vehicle attribute corresponding to the historical car rental order in the location area corresponding to the target car rental point (that is, the target car rental area) Information
  • the third vehicle attribute information of the vehicle to be dispatched may select the preset vehicle attribute information of the vehicle to be dispatched
  • the vehicle dispatch plan of the target car rental point can be determined through predictive model processing.
  • the location area corresponding to the target car rental spot is determined, and the location information belongs to the history of users in the location area.
  • the historical vehicle attribute information and the rental vehicle attribute information corresponding to the historical car rental orders are processed through the predictive model to obtain the associated vehicle demand of the target car rental point; then according to the vehicle attribute information of the vehicles to be dispatched, from the vehicles to be dispatched,
  • the target car rental point determines the dispatched vehicles and dispatches the determined vehicles to the target car rental point. In this way, the offline access rate of the target car rental point is improved, and the car rental efficiency of the target car rental point is also improved.
  • the user can directly rent the car at the target car rental point. Picking up the car without having to go to the central warehouse to see the car and pick up the car saves users time, improves user experience, and improves the service quality of the platform.
  • FIG. 10 shows an exemplary flowchart of a vehicle deployment method 1000 provided by some embodiments of the present application.
  • the method 1000 may be implemented by a processor.
  • Step 1001 Obtain location information of a target car rental point and historical vehicle attribute information of the vehicle rented at the target car rental point;
  • Step 1002 Determine a location area corresponding to the target car rental spot based on the location information of the target car rental spot;
  • Step 1003 Obtain historical car rental orders corresponding to the user terminal whose location information belongs to the location area;
  • Step 1004 Process the historical vehicle attribute information and the leased vehicle attribute information corresponding to the historical car rental order through a predictive model. According to the preset vehicle attribute information of the vehicle to be dispatched, it is determined from a plurality of vehicles to be dispatched.
  • the target car rental point determines the dispatch vehicle, and sends the determined dispatch vehicle information to the dispatch terminal, so that the determined vehicle can be dispatched to the target car rental point.
  • the target car rental point is the car rental point where the user can check the vehicle offline and pick up the car;
  • the location information can be expressed by the Global Positioning System (GPS) coordinates;
  • the target car rental point can be For the vehicles rented by the user at the target car rental point in the historical time period, for example, the car rented at the target car rental point by way of offline order, or the car rented through the car rental system of the target car rental point;
  • the information includes the license plate number of the vehicle, model information, vehicle driving information, vehicle maintenance information, etc., among which the model information can be a multi-purpose vehicle (MPV) model, a sport utility vehicle (Sport Utility Vehicle, SUV), Large cars, medium cars, small cars, etc.;
  • vehicle driving information includes vehicle mileage, for example, the vehicle has traveled 100 kilometers in total;
  • vehicle maintenance information includes vehicle annual inspection information, maintenance information, insurance information, etc.;
  • historical vehicle attribute information can be the target The car rental point is uploaded to the server.
  • the location area is an area including the target car rental point, that is, the area near the target shop rental point.
  • the location area may be a rectangular area, a circular area, or a pre-divided area.
  • the city can be divided according to administrative functions, or the location area can be obtained based on the surrounding information of the car rental point.
  • the location area corresponding to the target car rental point can be obtained by any of the following methods:
  • a circular area formed by taking the location information of the target car rental point as a center and a distance threshold as a radius is used as a location area corresponding to the target car rental point.
  • the preset distance threshold may be preset.
  • the distance threshold may be 5 kilometers. It should be noted here that the preset distance threshold may be determined according to actual conditions.
  • the location information of the target car rental point is taken as the center of the circle, and the distance threshold is used as the radius to form a circular area, and the circular area is taken as the location area corresponding to the target car rental point.
  • the location area corresponding to the target car rental point is determined.
  • the preset location area may be divided according to the administrative function of the area.
  • the location information range includes the location information of the target car rental location
  • the existing location information range is determined as the location area corresponding to the target car rental point.
  • the positioning information is the positioning location used by the user when browsing the vehicle on the car rental platform online.
  • the positioning information can be expressed by GPS coordinates; in the historical car rental order Including the vehicle attribute information of the leased vehicle by the user, such as the model information of the leased vehicle, vehicle driving information, vehicle maintenance information, etc.
  • vehicle attribute information of the leased vehicle by the user such as the model information of the leased vehicle, vehicle driving information, vehicle maintenance information, etc.
  • the vehicles to be dispatched are obtained by screening the vehicles of the vehicle supplier by the car rental platform.
  • the vehicles to be dispatched include vehicles corresponding to different models.
  • the car rental platform selects vehicles, it usually reports the number of vehicles through the vehicle supplier. , Vehicle attribute information and vehicle quotation.
  • the car rental platform determines the multiple models required and the number of vehicles corresponding to each model based on the vehicle quotation, model and other information; the vehicle attribute information of the vehicle to be dispatched includes the model of the vehicle to be dispatched, and the corresponding The total number of.
  • FIG. 11 shows an exemplary flowchart of a vehicle deployment method 1100 provided by some embodiments of the present application.
  • the method 1100 may be implemented by a processor.
  • Step 1101 Process the historical model information corresponding to each leased vehicle at the target car rental spot through a predictive model, and determine the first car model set corresponding to the target car rental spot, and the leased car models corresponding to different car models in the first car rental spot.
  • Step 1102 Process the model information of the leased vehicles included in each of the historical car rental orders through the predictive model, and determine the second model set corresponding to the location area and the second number of leased vehicles corresponding to different models in the second model set. ;
  • Step 1103 Process the first numbers corresponding to different car models in the first car model set and the second numbers corresponding to different car models in the second car model set through the predictive model, from the first car model set and the In the second car model set, determine the target car model, and determine the third number corresponding to the target car model;
  • Step 1104 Based on the vehicle attribute information of the vehicles to be dispatched, a third number of vehicles to be dispatched that match the target vehicle type are selected from the plurality of vehicles to be dispatched.
  • the first vehicle type set includes multiple vehicle types, and the first vehicle type set represents the vehicle type of the vehicle rented at the target car rental point. For example, there are 100 vehicles that have been rented at the target car rental point, of which 20 vehicles are of type A , 70 vehicles of model B and 10 vehicles of model C, then the models in the first model set are A, B, C; the first number is the number of vehicles corresponding to the models in the first vehicle set, for example, model A corresponds to 20 vehicles, Then the first number corresponding to vehicle type A is 20; the second vehicle type set is the vehicle type of the leased vehicle in the location area corresponding to the target car rental point, and the second number is the number of vehicles corresponding to the vehicle type in the second vehicle set.
  • the model corresponding to C1 is A1, the model corresponding to C2 is A2, the model corresponding to C3 is A1, and the model corresponding to C4 is The car model corresponding to A3 and C5 is A2, and the first car model set includes car models A1, A2, and A3.
  • the first number corresponding to car model A1 is 2, the first number corresponding to car model A2 is 2, and the first number corresponding to car model A3 is 2.
  • the number is 1.
  • the rental vehicles included in the historical car rental orders are C6, C7, C8, C9, and C10, respectively.
  • the car model corresponding to C6 is A1, and the car model corresponding to C7 is A5.
  • the model corresponding to C8 is A1, the model corresponding to C9 is A3, and the model corresponding to C10 is A1.
  • the second model set includes models A1, A3, and A5.
  • the first number corresponding to model A1 is 3, and the corresponding model A3 is The first number is 1, and the first number corresponding to model A5 is 1.
  • the preset number can be determined according to actual conditions.
  • the third number is the largest number of the first number and the second number corresponding to the same vehicle type. If the target vehicle type is only in the first vehicle type collection The third number is the first number corresponding to the model in the first model set. If the target model is only the model in the second model set, the third number is the corresponding model in the second model set. The second number.
  • the first car model set includes car models A1, A2, and A3, where the first number corresponding to car model A1 is 2, the first number corresponding to car model A2 is 2, the first number corresponding to car model A3 is 1, and the second car model collection It includes models A1, A3, and A5.
  • the first number corresponding to model A1 is 2, the first number corresponding to model A3 is 2, and the first number corresponding to model A5 is 1.
  • the first model set and the second model set are common
  • the included car models are A1 and A3, the number of leased vehicles corresponding to model A1 is 2, and the number of leased vehicles corresponding to A3 is 2, and the order of the models in descending order of number is A1 and A3, A2, A5, the preset number corresponding to the target car model is 3, then the target car model is A1, A3, A2, the third number corresponding to A1 is 2, the third number corresponding to A3 is 3, and the third number corresponding to A2 is 2 .
  • For each target model determined compare the target model with the model information of the vehicle to be dispatched. If the model information of the vehicle to be dispatched matches the target model, that is, the model information of the vehicle to be dispatched is consistent with the target model. Get the total number of vehicles that exist to be dispatched.
  • the third number corresponding to the target model is less than the total number of vehicles to be dispatched, the third number of vehicles to be dispatched is selected; if the third number corresponding to the target model is greater than the total number of vehicles to be dispatched, it can be the target car rental point Dispatch the total number of vehicles to be dispatched, calculate the difference between the third number and the total number of vehicles, wait for the central warehouse to have the number of vehicles to be dispatched, and then assign the target car rental point to the target car rental point. vehicle.
  • the model information of the vehicles to be dispatched includes A1, A2, A3, A4, and A5.
  • the total number of vehicles corresponding to model A1 is 10
  • the total number of vehicles corresponding to model A2 is 10
  • the total number of vehicles corresponding to model A3 is 10
  • the total number of vehicles corresponding to model A4 is 10.
  • the total number of corresponding vehicles is 10
  • the total number of vehicles corresponding to model A5 is 10
  • the target models include A1, A3, A2, the third number corresponding to A1 is 4, the third number corresponding to A3 is 3, and the third number corresponding to A2 is 2.
  • Assign 4 A1 type vehicles to the target car rental point 3
  • 3 type vehicles to the target car rental point and 2 A2 type vehicles to the target car rental point.
  • the dispatch vehicle required by the target car rental point may conflict with another car rental point.
  • the number of type A vehicles that need to be dispatched at the target car rental point is M
  • the other car rental point needs to be dispatched A type
  • the number of vehicles is N
  • the total number of vehicles to be dispatched for type A is less than M+N.
  • selecting the third number of vehicles to be dispatched that matches the target vehicle type from the plurality of vehicles to be dispatched It can include the following steps:
  • the model information matches the target model
  • the sum of the number of dispatched vehicles and the third number is greater than that corresponding to the target model
  • the total number of vehicles to be dispatched is compared with the priority of the target car rental point and another car rental point;
  • the third number of vehicles to be dispatched that match the target vehicle type are determined as the dispatch vehicles determined by the target car rental point.
  • the priority represents the importance of the car rental point. The higher the importance, the higher the priority of the car rental point.
  • the priority can be determined according to the resource value of the car rental point or the service value of the car rental point. It can be determined according to the actual situation.
  • the third number of vehicles to be dispatched that match the target model will be dispatched to the target car rental point; if the priority of the target car rental point is lower than that of another car rental point If the priority of the target car rental point is equal to the priority of the other car rental point, the vehicle can be allocated to the two car rental points on average, that is, the vehicle to be allocated to the target car model is calculated.
  • the ratio of the total number of dispatched vehicles to the number of car rental points, and each car rental point is allocated the same number of vehicles to be dispatched as the ratio.
  • the target car rental point S1 For example, four A1 vehicles are allocated to the target car rental point S1, and 5 A1 vehicles are allocated to another car rental point S2.
  • the priority of the target car rental point S1 is L1
  • the priority of another car rental point S2 is L2.
  • the common car model existing in the target car rental point S1 and another car rental point S2 is the A1 type
  • the target car rental point S1 and the other car rental point S2 have a total demand for A1 vehicles of 10
  • the A type is to be dispatched
  • the total number of vehicles is 8.
  • the total demand for A1 vehicles at the target car rental point S1 and another car rental point S2 is greater than the total number of A1 vehicles to be dispatched.
  • the vehicle demand information reported by the target car rental point or the user interest information of the target car rental point can also be considered to determine the vehicles that need to be dispatched for the target car rental point. The following are introduced separately.
  • the target is set from a plurality of vehicles to be dispatched
  • the car rental point determines the dispatched vehicle.
  • the vehicle demand information includes model information and the number of needs corresponding to each model.
  • the historical vehicle attribute information (model information) of the vehicles rented at the target car rental point is used to determine the first vehicle type set corresponding to the target car rental point and the first vehicle type corresponding to each vehicle type in the first vehicle type set.
  • the number based on the historical car rental orders of users in the location area where the target car rental point is located, determines the second vehicle type set corresponding to the location area where the target car rental point is located, and the second number corresponding to each vehicle type in the second vehicle type set.
  • the determination process of the first vehicle type set, the first number, the second vehicle type set, and the second number can be referred to the above, and will not be repeated here.
  • the target car rental point allocates the vehicles to be dispatched corresponding to the maximum number of models.
  • the target car rental point If only the vehicle type in the second vehicle type set matches any vehicle type required by the target car rental point, select the largest number from the number of vehicles corresponding to any vehicle type and the second number corresponding to the vehicle type in the second vehicle type set, and give The target car rental point allocates any vehicle type to be dispatched corresponding to the maximum number.
  • the model is randomly selected as the target model, and the number corresponding to the randomly selected model is used as the number of vehicles to be dispatched corresponding to the target model. For example, if the vehicle type A in the first vehicle type set is selected as the target vehicle type, the first number corresponding to the vehicle type A is determined as the number of vehicles of the type A vehicle that needs to be dispatched.
  • the target is set from a plurality of vehicles to be dispatched
  • the car rental point determines the dispatched vehicle.
  • the user interest information represents the vehicle type information that the user is interested in.
  • the user interest information may be determined based on the user's historical browsing data, or may be determined based on the user's marked interest, or may be determined according to actual conditions.
  • the process of determining the first number corresponding to each model in the first model set and the second model set, and the second number corresponding to each model in the second model set can be Refer to the above, and I won’t go into details here.
  • Obtain user interest information for visiting the target car rental location determine the user's interested model, compare the interested model, the models in the first model collection, and the models in the second model collection, if the first model collection is in the second model collection If the same car model exists, and the same car model matches the car model of interest, then the first number and the second number corresponding to the car model of interest are determined, the maximum number is selected, and the target car rental point is assigned the maximum number of waiting car models of interest. Dispatch vehicles.
  • the first number of vehicles to be dispatched of the vehicle type of interest are allocated to the target car rental point.
  • the second number of vehicles to be dispatched of the vehicle type of interest are allocated to the target car rental point.
  • the model is used as the number of vehicles to be dispatched corresponding to the target vehicle type. For example, if vehicle type A in the first vehicle type set is selected as the target vehicle type, the first number corresponding to vehicle type A is determined as the number of vehicles of type A vehicles that need to be dispatched. And record the interested models of the target points of interest for consideration in the next assignment.
  • the vehicle at the target car rental point Since the vehicle is in the process of dispatching to the target car rental point (that is, the vehicle is transported from the central warehouse to the target car rental point), there may be transportation problems that cause abnormal problems in the vehicles dispatched to the target car rental point, such as scratches on the car body, etc. Or the vehicle at the target car rental point has a problem of rental delay, that is, the vehicle at the target car rental point is not rented by the user within a certain period of time. In order to improve the rental efficiency of the target car rental point, the vehicle at the target car rental point can be called back.
  • the callback vehicle order includes historical service information of the callback vehicle
  • the target car rental point is instructed to control the callback vehicle to perform callback.
  • the historical service information includes information such as the license plate number, model information, number of vehicles, body information, annual inspection information, service life, mileage and other information of the recalled vehicle.
  • the number of vehicles is the number of vehicles corresponding to each model, and body information It is used to characterize whether the car body is scratched
  • the annual inspection information is used to characterize whether the vehicle is undergoing annual inspection
  • the service life characterizes the length of time the vehicle has been used
  • the driving distance characterizes the total number of kilometers traveled by the model.
  • the callback condition includes at least one of the following conditions:
  • the car body has scratches; the time since the annual inspection is less than the preset time; the service life is greater than the preset age threshold; the mileage is greater than the preset mileage threshold.
  • the preset duration may be one month
  • the preset age threshold may be determined according to the age of the vehicle being scrapped
  • the mileage threshold may be determined according to the historical vehicle mileage.
  • the target car rental point is instructed to callback the vehicle in the callback vehicle order to the vehicle recycling center.
  • selecting the closest vehicle recycling center for the target car rental point from multiple vehicle recycling centers may include the following steps:
  • the location information of the vehicle recycling center can be represented by GPS coordinates.
  • map information can be used to calculate the driving distance between the target car rental point and the vehicle recycling center, which is not limited in this application.
  • Completed Vehicle recycling facilitates the reuse of recycled vehicles and improves the utilization rate of recycled vehicles.
  • the target car rental point is S1
  • the vehicle recycling center is B1, B2, B3, the travel distance from S1 to B1 is L1
  • the travel distance from S1 to B2 is L2
  • the travel distance from S1 to B3 is L3, where L3 is the minimum travel Distance, it is determined that B3 is the vehicle recycling center that receives the callback vehicle, and the callback vehicle at the target rental point is transported to B3.
  • the embodiment of the application also provides a vehicle deployment device corresponding to the vehicle deployment method. Since the principle of the method in the embodiment of the application to solve the problem is similar to the above-mentioned vehicle deployment method in the embodiment of the application, the implementation of the device can be Refer to the implementation of the method, and the repetition will not be repeated.
  • the device 1200 includes:
  • the first obtaining module 1201 is configured to obtain the location information of the target car rental point and the historical vehicle attribute information of the vehicles rented at the target car rental point;
  • the determining module 1202 is configured to determine a location area corresponding to the target car rental point based on the location information of the target car rental point;
  • the second acquiring module 1203 is configured to acquire historical car rental orders corresponding to the user terminal whose location information belongs to the location area;
  • the processing module 1204 is configured to process the historical vehicle attribute information and the rental vehicle attribute information corresponding to the historical car rental order through a predictive model, and according to preset vehicle attribute information of the vehicles to be dispatched, from a plurality of vehicles to be dispatched , Determine the dispatch vehicle for the target car rental point, and send the determined dispatch vehicle information to the dispatch terminal, so that the determined vehicle can be dispatched to the target car rental point.
  • the acquiring module 501 can acquire the location information of the target car rental point and the historical vehicle attribute information of the vehicle rented at the target car rental point, determine the location area corresponding to the target car rental point, and obtain the positioning information belonging to the target car rental point.
  • the acquiring module 501 may include multiple sub-modules.
  • the acquiring module 501 may include a first acquiring module 1201, a determining module 1202, and a second acquiring module 1203.
  • the processing module 1204 here may include the aforementioned first determining module 502 and the second determining module 503.
  • the processing module 1204 determines the vehicles that need to be dispatched for the target car rental point from a plurality of vehicles to be dispatched according to the following steps:
  • the historical model information corresponding to each leased vehicle at the target car rental point is processed by a predictive model, and the first model set corresponding to the target car rental point is determined, and the first vehicle model set corresponding to different models in the first model set is determined.
  • the first number corresponding to different models in the first model set and the second number corresponding to different models in the second model set are processed by a predictive model, from the first model set and the second model set. In the collection, determine the target vehicle type, and determine the third number corresponding to the target vehicle type;
  • a third number of vehicles to be dispatched that match the target vehicle type are selected from the plurality of vehicles to be dispatched.
  • processing module 1204 is further configured to:
  • the total number of vehicles to be dispatched is compared with the priority of the target car rental point and another car rental point;
  • the third number of vehicles to be dispatched that match the target vehicle type are determined as the dispatch vehicles determined by the target car rental point.
  • processing module 1204 is specifically configured to:
  • the target is set from a plurality of vehicles to be dispatched
  • the car rental point determines the dispatched vehicle.
  • processing module 1204 is specifically configured to:
  • the target is The car rental point determines the dispatched vehicle.
  • the first obtaining module 1201 is further used for:
  • the callback vehicle order includes historical service information of the callback vehicle
  • the determining module 1202 is also used for:
  • the target car rental point is instructed to control the callback vehicle to perform callback.
  • the determining module 1202 is further configured to instruct the target car rental point to control the callback vehicle for callback according to the following steps:
  • the determining module 1202 is specifically configured to:
  • a circular area formed by taking the location information of the target car rental point as a center and a distance threshold as a radius is used as a location area corresponding to the target car rental point.
  • An embodiment of the present application also provides an electronic device 1300.
  • a schematic structural diagram of the electronic device 1300 provided in an embodiment of the present application includes: a processor 1301, a memory 1302, and a bus 1303.
  • the memory 1302 stores machine-readable instructions executable by the processor 1301 (for example, execution instructions corresponding to the first acquiring module 1201, the determining module 1202, the second acquiring module 1203, and the processing module 1204 in the device in FIG. 12 Etc.), when the electronic device 1300 is running, the processor 1301 and the memory 1302 communicate through the bus 1303, and the machine-readable instructions are executed by the processor 1301 when the following processing is performed:
  • the instructions executed by the processor 1301 are based on the historical vehicle attribute information and the rental vehicle attribute information corresponding to the historical car rental order, and the preset vehicle attribute information of the vehicle to be dispatched.
  • determining the dispatching vehicle for the target car rental point includes:
  • a third number of vehicles to be dispatched that match the target vehicle type are selected from the plurality of vehicles to be dispatched.
  • a third number of vehicles that match the target vehicle type are selected from the plurality of vehicles to be dispatched.
  • Vehicles to be dispatched including:
  • the model information matches the target model
  • the sum of the number of dispatched vehicles and the third number is greater than that corresponding to the target model
  • the total number of vehicles to be dispatched is compared with the priority of the target car rental point and another car rental point;
  • the third number of vehicles to be dispatched that match the target vehicle type are determined as the dispatch vehicles determined by the target car rental point.
  • determining a dispatch vehicle for the target car rental point from a plurality of vehicles to be dispatched includes:
  • the target is set from a plurality of vehicles to be dispatched
  • the car rental point determines the dispatched vehicle.
  • determining a dispatch vehicle for the target car rental point from a plurality of vehicles to be dispatched includes:
  • the target is The car rental point determines the dispatched vehicle.
  • the instructions executed by the processor 1301 further include:
  • the callback vehicle order includes historical service information of the callback vehicle
  • the target car rental point is instructed to control the callback vehicle to perform callback.
  • the instruction executed by the processor 1301 instructing the target car rental point to control the callback vehicle for callback includes:
  • the instruction executed by the processor 1301 to determine the location area corresponding to the target car rental point based on the location information of the target car rental point includes:
  • a circular area formed by taking the location information of the target car rental point as a center and a distance threshold as a radius is used as a location area corresponding to the target car rental point.
  • the embodiment of the present application also provides a computer-readable storage medium having a computer program stored on the computer-readable storage medium, and the computer program executes the steps of the vehicle deployment method described above when the computer program is run by a processor.
  • the storage medium can be a general storage medium, such as a mobile disk, a hard disk, etc., when the computer program on the storage medium is run, the above-mentioned vehicle scheduling method can be executed, which is used to solve the problem of low car rental efficiency at car rental points in the prior art. problem.
  • the vehicle deployment method and its hardware device can be applied to the scenario of a car rental platform.
  • the server of the target car rental point can first determine its target car rental area information, and send the location information to the back-end server of the car rental platform based on the target car rental area information to obtain the data acquisition request of the user’s historical car rental order in the target car rental area.
  • the end server can respond to the data acquisition request and transmit corresponding data to the server of the target car rental point; the server of the target car rental point receives the historical car rental order data of the user in the target car rental area and the historical vehicle attribute information of the vehicle rented at the target car rental point , And then can use its pre-stored prediction model to perform data processing to obtain dispatched vehicle information; according to the dispatched vehicle information, the server of the target car rental point can send the corresponding vehicle dispatch request containing the vehicle dispatch information to the server of the corresponding central warehouse, the server of the central warehouse Carry out corresponding vehicle scheduling operations according to the vehicle scheduling request, and at the same time update the data of the server of the target car rental point and the data update of the car rental platform after the vehicle scheduling process and completion.
  • the vehicle type situation (model model, remaining number of vehicle models, etc.) corresponding to the target car rental location in the target car rental area can be obtained through query Target car rental point address, etc.)
  • the model details interface After selecting the car model, enter the model details interface, and you can choose the one that suits your own according to the provided car rental plan to place an order or conduct consultations on the target car rental spot (such as the corresponding offline store) according to the provided car rental plan.
  • the short-term rental type interface includes the pick-up point and time, and then click to select the car, enter the car selection list and so on.
  • the location area corresponding to the target car rental spot is determined, and the location information belongs to the historical car rental order of the user in the location area.
  • the dispatched vehicle is determined for the target car rental point, and the information of the determined dispatched vehicle is sent to The dispatch terminal can dispatch the determined vehicle to the target car rental point.
  • the offline access rate of the target car rental point is improved, and the car rental efficiency of the target car rental point is also improved.
  • the user can directly mention the car at the target car rental point. It saves users time, improves user experience, and improves the service quality of the platform without having to go to the central warehouse to see the car and pick up the car.
  • FIG. 14 is an exemplary flowchart of a vehicle deployment method 1400 according to some embodiments of the present application.
  • the method 1400 may be executed by a processor.
  • Step 1401 Obtain the associated feature data of the target car rental location, where the associated feature data includes: environmental information of the target car rental location and product selection related feature data of the target car rental area;
  • Step 1402 based on the prediction model, process the environmental information of the target car rental location and the product-related feature data of the target car rental area, and determine the model demand of the target car rental location;
  • Step 1403 Determine a vehicle scheduling plan for the target car rental point based at least on the associated vehicle demand.
  • the vehicle allocation method provided in this embodiment analyzes the environmental information of the target car rental location and the product selection related characteristics of the target car rental area through a predictive model, and determines the vehicle type demand of the target car rental location. This method saves the labor cost invested in manual sorting and screening, and the model demand determined by the predictive model can effectively fit the target car rental area environment and the car rental demand in the local market, thereby improving the car rental efficiency.
  • FIG. 15 is an exemplary flowchart of a vehicle deployment method 1500 according to some embodiments of the present application.
  • the method 1500 may be executed by a processor.
  • Step 1501 Obtain the associated feature data of the target car rental location, where the associated feature data includes: environmental information of the target car rental location and product selection related feature data of the target car rental area;
  • Step 1502 based on the prediction model, process the environmental information of the target car rental location and the product selection-related feature data of the target car rental area, and determine the model demand of the target car rental location;
  • Step 1503 based on the third vehicle attribute information of the vehicles to be dispatched and the model requirements of the target car rental point, determine the dispatched vehicles of the vehicle dispatch plan from the vehicles to be dispatched.
  • FIG. 16 is an exemplary flowchart of a vehicle deployment method 1600 according to some embodiments of the present application.
  • the method 1600 may be executed by a processor.
  • Step 1601 Obtain associated feature data of the target car rental location, where the associated feature data includes: environmental information of the target car rental location and data related to historical product selection requirements in multiple dimensions of the target car rental area;
  • Step 1602 Based on the prediction model, process the environmental information of the target car rental location and the multiple dimensions of historical product selection demand related data in the target car rental area, and determine the target car model of the target car rental spot and the number of corresponding target car models;
  • Step 1603 based on the third vehicle attribute information of the vehicles to be dispatched and the target car model of the target car rental site and the number of corresponding target car models, select a dispatched vehicle matching the target car model of the target car rental site and the number of target car models.
  • the vehicle deployment method provided in this embodiment uses a predictive model to process the environmental information of the target car rental location and the historical selection demand related data in multiple dimensions of the target car rental area to determine the matching target vehicle model and the corresponding number of the target car rental location, and Based on this, the dispatch vehicle is selected. While saving labor costs, this method can also take into account realistic factors such as market changes and vehicle supplier dynamics, timely and effectively meet the environment of the target car rental area and the needs of the local car rental market, and improve the efficiency of car rental.
  • the environmental information of the target car rental location may select the environmental information of each central warehouse in the target city.
  • the historical selection requirements data of multiple dimensions in the target car rental area include the characteristic data of each central warehouse and the information of the target city.
  • the third vehicle attribute information of the vehicle to be dispatched can select the preset vehicle attribute information of the vehicle to be dispatched, and the vehicle dispatch plan of the target car rental point can be determined through predictive model processing.
  • FIG 17 is a schematic diagram of an application scenario in which the vehicle deployment method and its hardware device are applied to a user using a car rental platform.
  • this scenario includes: a user and a car rental platform; among them, the user can open the car rental function of the car rental platform on his own terminal device, and the user can select the type of car rental in the car rental interface, such as long-term rental, short-term rental, Car sharing.
  • Different car rental types correspond to different car selection scenarios. For example, on the long-term rental type interface, you can directly select the car model to be rented. After selecting the car model, you can enter the model details interface.
  • the car rental platform can provide users with a variety of car rental solutions and provide a variety of car rental models. Therefore, in order to improve the matching degree with user needs, the selection of the central warehouse on the car rental platform is particularly important. The higher the accuracy, The higher the matching degree, the higher the vehicle turnover rate of the central warehouse.
  • the following vehicle deployment method aims to solve the above technical problems of the prior art.
  • FIG. 18 shows an exemplary flowchart of a vehicle deployment method 1800 according to some embodiments of the present application.
  • the method 1800 may be implemented by a processor.
  • Step 1801 Obtain characteristic data of the target city in the first preset historical time period.
  • the target city is a city with at least one central warehouse, and the characteristic data is used to represent the environmental information of each central warehouse in the target city and the associated data of the vehicle type in the target city.
  • the execution body may be a server of the car rental platform.
  • the server obtains characteristic data of a certain city in a preset historical time period, for example, characteristic data corresponding to the city one month ago.
  • the city is regarded as a target city below, where the target city has at least one central warehouse, which is used to store vehicles of various models, and the vehicles in the central warehouse are dynamically derived from various leasing companies.
  • the characteristic data corresponding to the target city includes the characteristic data of each central warehouse and the unique characteristics of the target city.
  • the characteristic data of the central warehouse itself may include the environmental information of the location of the central warehouse, such as the population density of the area in the target city that the central warehouse can cover, road characteristics, such as the number of intersections, the shape, and whether the traffic is convenient or not.
  • the unique characteristics of the target city can include information about the objective environment of the entire city, vehicle information related to the objective environment, such as the population density of the entire city, road characteristics, and vehicle-related information (for example, the type of vehicle that is ordered on the platform, the The city's preferred vehicle type, etc.). Therefore, the characteristic data corresponding to the target city has the characteristics of the city or each selected central warehouse, and the multi-influencing factors of the characteristic data provide an effective data source for the accuracy of the central warehouse selection.
  • step 1802 the characteristic data is processed through the prediction model to obtain the vehicle demand prediction data corresponding to each central warehouse in the target city within a preset time period.
  • the prediction model is obtained by training the decision tree model.
  • each central warehouse since the selection of each central warehouse directly affects the vehicle turnover efficiency of the central warehouse, in order to improve the turnover efficiency, the demand for vehicle models in the target city in a period of time in the future can be predicted, and then the vehicle models and the central warehouse can be adjusted in time. Quantity.
  • the decision tree algorithm is used to train the decision tree model through historical feature data to obtain the prediction model, and then the current feature data of the target city, such as the feature data of the previous month, is passed through the data
  • the forecasting model it is input into the forecasting model to obtain the forecasted value of the demand for the vehicle model in each central warehouse in the target city within a preset time period (for example, one week in the future), that is, the forecast data of vehicle demand.
  • the forecast is the demand for models in the future, such as the demand for models in the next week
  • the previous historical data including the data at the current time node, such as one month
  • the forecast method can be a dynamic iterative of daily updates, realizing the refined selection of the central warehouse.
  • the input data of this forecasting method is characteristic data that takes into account multiple factors, it is more in line with the needs of local residents. Therefore, the determination of the demand forecast data of this vehicle model can produce a higher degree of matching between the selection of the central warehouse and the needs of users.
  • step 1802 is described in detail.
  • obtaining the forecast data of vehicle demand corresponding to each central warehouse in the target city within a preset time period through a prediction model includes:
  • Step a1 normalizing the feature data to obtain the target feature quantity corresponding to each central warehouse in the target city where the feature data is distributed;
  • Step a2 Input the feature quantity corresponding to the target feature data into the prediction model to obtain the model demand forecast data of each central warehouse in the target city within a preset time period.
  • the feature data of the target city is first normalized, that is, feature quantization is performed, and converted into digital features. For example, converting language features into digital vectors. Then the feature data of the target city is integrated into at least one piece of data, each piece of data includes the feature quantity corresponding to a central warehouse and the inherent feature quantity corresponding to the target city (such as the above-mentioned population density of the entire city, road characteristics, and vehicle-related Information, etc.), and then each piece of data is input into the prediction model, and the model demand forecast data of each central warehouse in the target city within a preset time period can be obtained.
  • the prediction result integrates the characteristics of each central warehouse in the target city and factors such as the road, population, and car model preference in the target city. Therefore, the determination of the demand forecast data for this vehicle model is not only accurate, but also makes the central warehouse parked. The vehicle can match the user's needs with a high degree, thereby increasing the vehicle turnover rate.
  • Step 1803 Determine the product selection result of each of the central warehouses according to the forecast data of the vehicle model demand.
  • each central warehouse can be assigned to The model demand forecast data is used as the basis for product selection. For example, the models in the user area covered by the central warehouse are sorted according to the size of the forecast data, and the top models in the ranking are finally selected as the product selection result of the central warehouse.
  • the vehicle deployment method provided in the above embodiment realizes the connection of full-link data through the acquisition of environmental information of each central warehouse in the target city and the associated data of the vehicle model in the target city, taking into account multiple factors, and combining with machine learning algorithms For example, by training the decision tree model to achieve vehicle demand prediction without manual intervention, it also greatly saves human resources, and then based on the forecast data, the selection of the central warehouse is made to make a precise selection, and the central warehouse is parked The high degree of matching between the vehicles and the user’s needs enables the vehicles parked in the central warehouse to have a high degree of matching with the user’s needs, which improves the vehicle turnover rate.
  • the vehicle demand forecast data includes the forecast demand of multiple pre-selected vehicle models.
  • the determining the product selection result of each of the central warehouses according to the demand forecast data of the vehicle type includes:
  • Step b1 According to the predicted demand of each pre-selected vehicle model among the multiple pre-selected vehicle models, the first actual demand of each central warehouse in the target city for each pre-selected vehicle model within a preset time period is calculated through the demand conversion rate.
  • Step b2 for each central warehouse, sort the first actual demand corresponding to each of the pre-selected vehicle types according to the order from high to low;
  • step b3 all pre-selected car models in the preset ranking order are used as the product selection results of the central warehouse.
  • the model demand forecast data corresponding to each central warehouse includes multiple pre-selected models and the demand for each pre-selected model.
  • the predicted future preset time period (for example, one week in the future) is predicted for each model.
  • the demand for each pre-selected model is more in line with the needs of the target city coverage area where the central warehouse is located, that is, it has a high degree of matching with user needs, and due to the local environment and scale limitations of the central warehouse, it is necessary to choose from multiple pre-selected models
  • the final model Therefore, the final selected model can be determined based on the demand for the pre-selected model in the model demand forecast data corresponding to each central warehouse.
  • the demand conversion rate is obtained by comparing the historical demand with the predicted value corresponding to the historical demand; the predicted value corresponding to the historical demand is obtained during the training process of the decision tree model.
  • the product selection result takes the top ten pre-selected car models in sequence as the selected car models of the central warehouse, and the product selection result includes the demand for each selected car model. In this way, the current model situation in the central warehouse can be reasonably supplemented or dispatched according to the selected model and the demand.
  • FIG. 19 is an exemplary flowchart of a vehicle deployment method 1900 according to an embodiment of the present application. This embodiment is based on the foregoing embodiment, for example, in FIG. On the basis of the embodiment described in 18, how to establish a prediction model is described in detail.
  • the method Before obtaining the forecast data of vehicle model demand corresponding to each central warehouse in the target city within a preset time period, the method further includes:
  • Step 1901 Obtain historical feature data of each of the multiple predetermined cities in the second preset historical time period and the central warehouses of each of the predetermined cities in the third preset historical time period for each vehicle model.
  • the historical demand of, the historical characteristic data includes data of multiple dimensions;
  • Step 1902 Train the decision tree model according to the data of the multiple dimensions and the historical demand corresponding to each of the predetermined cities to obtain the prediction model.
  • the time interval of the first preset history time period is equal to the time interval of the second preset history time period
  • the time interval of the second preset history time period is greater than the time interval of the third preset history time period
  • the preset time period The time interval is equal to the time interval of the third preset historical time period.
  • the first preset historical time period can be one month before the current time, such as 2019.10-2019.11, and the time interval is one month;
  • the second preset historical time period can be one or multiple, and can be combined with The time on the first preset historical time period is crossed, and the end time of the second preset historical time period can also be before the start time of the first preset historical time period (for example, 2019.9-2019.10), and the time interval is one Month;
  • the start time of the third preset historical time period can be the end time of the second preset historical time period (for example, 2019.10.1-2019.10.7), and the time interval of the third preset historical time period is 7 days That is, one week, the start time of the preset time period may be the end time of the first preset historical time period (for example, 2019.11.1-2019.11.7), and the time interval of the preset time period is 7 days.
  • each predetermined city is provided with one or more central warehouses.
  • the multiple predetermined cities may or may not include the target city, which is not limited here.
  • the historical characteristic data of each predetermined city in a second preset historical time period includes the inherent characteristic data of the predetermined city and the characteristic data of each central warehouse in the predetermined city. Therefore, a second predetermined historical time period
  • the historical feature data within can include multiple dimensions of data. Then, according to the data of multiple dimensions and the historical demand corresponding to each central warehouse in the predetermined city as training data, the decision tree model is trained to obtain the prediction model.
  • multiple dimensions can include: car rental platform order dimension, car rental platform bubbling dimension, car rental platform driver dimension, urban travel dimension, central warehouse environment dimension, urban population density dimension, urban environment dimension, urban rental company dimension, and urban population Income dimension.
  • the data of the order dimension of the car rental platform represents the number of car models ordered in the central warehouse
  • the data of the car rental platform bubbling dimension represents the number of models that the user has selected on the car rental platform but has not placed an order
  • the driver dimension of the car rental platform The data of represents the number of drivers in the predetermined city on the car rental platform
  • the data of the urban travel dimension represents the monthly average historical travel data volume of all the population of the predetermined city
  • the data of the central warehouse environment dimension represents the area covered by the central warehouse The density of the population and the characteristics of the roads around the central warehouse.
  • the data of the urban population density dimension represents the population density of the predetermined city
  • the data of the urban environment dimension represents the road characteristics of the predetermined city
  • the data of the city leasing company dimension represents the The word of mouth and scale of all leasing companies in the scheduled city
  • the data of the urban population income dimension represents the average monthly income of all the population in the scheduled city.
  • these dimensions take into account the number of car models ordered on the car rental platform in the city where they are located and the user's behavior data on the car rental platform (for example, the number of car models that the user has selected on the car rental platform but did not place an order), as well as the city where they are located
  • the population density, traffic environment, income, and the specific population density and traffic environment of the area covered by each central warehouse are also considered.
  • the reputation and scale of the leasing company that selects the model to enter the central warehouse, etc. may lead to the forecast of vehicle demand Multi-factors of data accuracy, using multi-factor data as training data to train the decision tree improves the accuracy of the parameters, makes the prediction model obtained after training better, and thus ensures the accuracy of the prediction result.
  • FIG. 20 is a schematic flowchart of a vehicle deployment method 2000 according to some embodiments of the present application. How to train the decision tree model based on the multiple dimensions of data and the historical demand corresponding to each of the predetermined cities to obtain the prediction model can be achieved through the following steps:
  • Step 2001 Perform normalization processing on the data of the multiple dimensions to obtain feature quantities of the multiple dimensions.
  • pre-defined coding is performed on the data of the central warehouse environment dimension, the data of the urban environment dimension, and the data of the city leasing company dimension to obtain the characteristic quantity of the central warehouse environment dimension and the urban environment
  • the feature quantity of the dimension and the feature quantity of the city rental company dimension is performed on the data of the central warehouse environment dimension, the data of the urban environment dimension, and the data of the city leasing company dimension to obtain the characteristic quantity of the central warehouse environment dimension and the urban environment
  • the feature quantity of the dimension and the feature quantity of the city rental company dimension is performed on the data of the central warehouse environment dimension, the data of the urban environment dimension, and the data of the city leasing company dimension to obtain the characteristic quantity of the central warehouse environment dimension and the urban environment
  • the feature quantity of the dimension and the feature quantity of the city rental company dimension is performed on the data of the central warehouse environment dimension, the data of the urban environment dimension, and the data of the city leasing company dimension.
  • a central warehouse corresponds to a characteristic quantity of the environmental dimension of the central warehouse.
  • the number of drivers in the driver dimension of the car rental platform is taken as the characteristic quantity of the driver dimension of the car rental platform
  • the monthly average number of trips in the city travel dimension is taken as the characteristic quantity of the city travel dimension
  • the urban population density dimension is The population density value is used as the characteristic quantity of the urban population density dimension
  • the per capita monthly income value of the urban population income dimension is used as the characteristic quantity of the urban population income dimension.
  • step 2002 the decision tree model is trained by taking the feature quantities of multiple dimensions corresponding to each central warehouse of each predetermined city and the historical demand as a training sample, wherein the historical demand is Is the label in the training process of the decision tree model;
  • Step 2003 according to the output of the decision tree model and the difference between the historical demand as the label, adjust the parameters of the decision tree model until the decision tree model achieves a desired training effect;
  • step 2004 the decision tree model that achieves the desired training effect is used as the prediction model.
  • the characteristics of multiple dimensions of each predetermined city and the historical demand corresponding to each of the multiple central warehouses can be divided into multiple samples , Where the number of samples is consistent with the number of central warehouses in the predetermined city.
  • sample 11 includes the feature quantity of the central warehouse 11 environment dimension, the feature quantity of the urban environment dimension, the feature quantity of the city leasing company dimension, the feature quantity of the order dimension of the car rental platform, and the feature quantity of the bubbling dimension of the car rental platform
  • the training sample is the input of the model
  • the historical demand is the label in the training process of the decision tree model
  • the output is the model demand forecast data. Then the output is compared with the label, the error is calculated, and the The parameters of the decision tree model are adjusted by feedback until the error between the output and the label is small and stable, that is, the decision tree model reaches the desired training effect, and the trained decision tree model is used as the prediction model.
  • step 2002 is described in detail.
  • the training of the decision tree model by taking the feature quantities of multiple dimensions corresponding to each central warehouse of each predetermined city and the historical demand as a training sample includes the following steps:
  • Step c1 Generate a first matrix according to the feature quantity of each dimension corresponding to each central warehouse of each predetermined city;
  • Step c2 Generate a second matrix according to the historical demand of each vehicle type corresponding to each central warehouse of each predetermined city;
  • Step c3. Form the training sample according to the first matrix and the second matrix, where the training sample is a combined matrix of the first matrix and the second matrix, wherein the first matrix is The first input quantity X in the merge matrix, the second matrix is used as the label input quantity Y of the merge matrix, and the first input quantity X corresponds to the unique label input quantity Y;
  • Step c4 Input the first input quantity X and the label input quantity Y into the decision tree model for training simultaneously, and output the predicted value corresponding to the label input quantity Y.
  • the characteristic quantities of each of the dimensions corresponding to each central warehouse include the characteristic quantities of the central warehouse environment dimension, the characteristic quantities of the urban environment dimension, the characteristic quantities of the city rental company dimension, and the dimension of the order placed on the car rental platform.
  • the first matrix generated by the feature quantity of each dimension is [the feature quantity of the central warehouse environment dimension, the feature quantity of the urban environment dimension, the feature quantity of the city rental company dimension, the feature quantity of the car rental platform, the single dimension of the car rental platform, the bubbling dimension of the car rental platform
  • the characteristic quantity of the driver dimension of the car rental platform The characteristic quantity of the urban travel dimension
  • the characteristic quantity of the urban population density dimension The characteristic quantity of the urban population income dimension]
  • the second matrix is generated by the historical demand of each vehicle
  • FIG. 21 is an exemplary flowchart of a vehicle deployment method 2100 according to some embodiments of the present application.
  • the method for selecting products in the central warehouse is described in detail.
  • the method may further include:
  • Step 2101 Push the car model corresponding to the product selection result to each user terminal, so that users of each user terminal provide feedback on the car model corresponding to the product selection result;
  • Step 2102 Receive feedback information from each of the user terminals, and adjust the product selection result according to the feedback information.
  • FIG. 22 is an example diagram of an application scenario of a vehicle deployment method 2100 according to some embodiments of the present application.
  • the server may push the calculated product selection result, that is, the desired car model, to each user terminal through the car rental platform.
  • the push message of the car rental platform reads: "Please rate the following models: 5 points, 3 points, 1 point”
  • the user selects the corresponding scores according to their preferences through the user terminal, and then clicks " "Submit”
  • the user terminal will feed back the results submitted by the user to the server in the car rental platform, and the server will adjust the demand for each model in the selection result according to the feedback, so that the adjusted selection result can make the user have a higher Satisfaction.
  • the vehicle deployment method also includes:
  • each of the central warehouses calculates the number of each vehicle type to be replenished in each of the central warehouses; or, according to the corresponding adjustments of each of the central warehouses Calculate the number of each vehicle type to be added for each central warehouse based on the result of the product selection and the current number of each vehicle type in each of the central warehouses.
  • the two calculation methods are to calculate the difference between the data of the current model in the center warehouse and the demand of the corresponding model in the product selection result, and use the difference as the number of each model to be filled in the center warehouse.
  • other algorithms can also be used to supplement the number of various models in the central warehouse to ensure the subsequent dispatch of various models of vehicles in the central warehouse.
  • the training samples are stored in a sample library; in order to dynamically update the samples or optimize the prediction model, after the determination of the product selection results of each of the central warehouses, the method further includes:
  • the true value of the demand for vehicle models of each central warehouse in the target city within the preset time period that is, the second The actual demand
  • the real value with the first actual demand predicted by the forecast model and use the multi-dimensional feature quantity corresponding to each central warehouse in the target city and the corresponding second actual demand as the new The samples are stored in the database so that the database can be dynamically updated.
  • the demand conversion rate can be updated, thereby improving the accuracy of the product selection result. That is, after the product selection result of each of the central warehouses is determined, the method may further include:
  • the true value of the demand for vehicle models of each central warehouse in the target city within the preset time period that is, the second actual demand amount
  • the true value of the demand for vehicle models of each central warehouse in the target city within the preset time period that is, the second actual demand amount
  • the vehicle deployment method provided in the above embodiments can be dynamically iteratively updated daily during the entire process. It is compared daily based on the information of the vehicles parked in the central warehouse and the calculated product selection results of the central warehouse, and automated decision-making is made. It is also necessary for the vehicle provider (such as a leasing company) to provide which vehicle type and how many vehicles each provide. This solution realizes the refined selection of the central warehouse and is updated daily. Through the full link data connection, End-to-end forecasting and decision-making are realized without manual intervention, which greatly saves operating manpower and enables more precise product selection, so that the vehicles parked in the central warehouse can match the needs of users and improve the vehicle turnover rate.
  • the vehicle provider such as a leasing company
  • FIG. 23 is an exemplary structure diagram of a vehicle deployment device 2300 according to some embodiments of the present application.
  • the vehicle deployment device may specifically be the car rental platform in the foregoing embodiment.
  • the vehicle deployment device can execute the processing flow provided by the vehicle deployment method embodiment. As shown in FIG.
  • the vehicle deployment device 2300 includes: a third acquisition module 2301, a prediction module 2302, and a product selection module 2303; among them, the third acquisition module 2301 , Used to obtain characteristic data of a target city in a first preset historical time period, where the target city is a city with at least one central warehouse, and the characteristic data is used to indicate each central warehouse in the target city The environment information and the associated data of the vehicle type in the target city; the prediction module 2302 is used to obtain the corresponding vehicle demand forecast for each central warehouse in the target city within a preset time period through a prediction model based on the characteristic data Data, the prediction model is obtained by the decision tree model training; the product selection module 2303 is used to determine the product selection results of each of the central warehouses according to the model demand prediction data.
  • the third acquisition module 2301 Used to obtain characteristic data of a target city in a first preset historical time period, where the target city is a city with at least one central warehouse, and the characteristic data is used to indicate each central warehouse in the target city The environment information
  • the acquisition module 501 may be used to acquire characteristic data of a target city in a first preset historical time period, where the target city is a city with at least one central warehouse, and the characteristic data is used to indicate The environmental information of each central warehouse in the target city and the associated data of the vehicle type in the target city.
  • the acquiring module 501 may include a third acquiring module 2301.
  • the aforementioned first determination module 502 may include a prediction module 2302.
  • the aforementioned second determining module 503 may include a product selection module 2303.
  • the device 2300 further includes: a model establishment module 2304, and the model establishment module 2304 includes: a first acquisition unit and a model training unit.
  • the first acquiring unit is configured to acquire historical feature data of each predetermined city in the second preset historical time period in a plurality of predetermined cities and the central warehouse of each predetermined city in the third predetermined historical time period
  • the historical demand for each model in the vehicle, and the historical feature data includes data of multiple dimensions.
  • the model determining unit is configured to train the decision tree model to obtain the prediction model according to the data of the multiple dimensions corresponding to each of the predetermined cities and the historical demand; wherein, the first preset The time interval of the historical time period is equal to the time interval of the second preset historical time period, the time interval of the second preset historical time period is greater than the time interval of the third preset historical time period, and the time interval of the preset time period is equal to the third The time interval of the preset historical time period.
  • the multiple dimensions include: car rental platform ordering dimension, car rental platform bubbling dimension, car rental platform driver dimension, urban travel dimension, central warehouse environment dimension, urban population density dimension, urban environment dimension, and urban rental company dimension And the urban population income dimension;
  • the model determination unit includes: a data processing subunit, a model training subunit, a parameter adjustment subunit, and a model determination subunit.
  • the data processing subunit is used to perform normalization processing on the data of the multiple dimensions to obtain feature quantities of the multiple dimensions.
  • the model training subunit is used to train the decision tree model with the feature quantities of multiple dimensions corresponding to each central warehouse of each predetermined city and the historical demand as a training sample, wherein The historical demand is the label in the training process of the decision tree model.
  • the parameter adjustment subunit is used to adjust the parameters of the decision tree model according to the output of the decision tree model and the difference between the historical demand as the label, until the decision tree model reaches the desired training effect.
  • the model determination subunit is used to use the decision tree model that achieves the desired training effect as the prediction model.
  • the data processing subunit is specifically configured to: perform predefined coding on the data of the central warehouse environment dimension, the data of the urban environment dimension, and the data of the city rental company dimension to obtain the central warehouse environment
  • the feature quantity of the dimension, the feature quantity of the urban environment dimension, and the feature quantity of the urban rental company dimension ; taking the number of car models ordered in the order dimension of the car rental platform as the characteristic quantity of the order dimension of the car rental platform,
  • the bubbling number of users of the car rental platform bubbling dimension is taken as the characteristic quantity of the car rental platform bubbling dimension
  • the number of drivers of the car rental platform driver dimension is taken as the characteristic quantity of the driver dimension of the car rental platform.
  • the monthly average number of trips of the dimension is taken as the characteristic quantity of the urban travel dimension
  • the population density value of the urban population density dimension is taken as the characteristic quantity of the urban population density dimension
  • the per capita monthly income value of the urban population income dimension is taken as the characteristic quantity of the income dimension of the urban population.
  • the model training subunit is specifically configured to: generate a first matrix according to the feature quantities of each of the dimensions corresponding to each central warehouse of each of the predetermined cities;
  • the historical demand of each vehicle model corresponding to the central warehouse is used to generate a second matrix;
  • the training samples are formed according to the first matrix and the second matrix, and the training samples are the first matrix and the second matrix.
  • a matrix of the merging matrix wherein the first matrix is the first input quantity X in the merging matrix, the second matrix is used as the label input quantity Y of the merging matrix, and the first input quantity X corresponds to the unique
  • the label input quantity Y; the first input quantity X and the label input quantity Y are synchronously input into the decision tree model for training, and the predicted value corresponding to the label input quantity Y is output.
  • the prediction module is specifically configured to: normalize the characteristic data to obtain the target characteristic quantity corresponding to each central warehouse in the target city; The feature quantity corresponding to the data is input into the prediction model, and the model demand forecast data of each central warehouse in the target city within a preset time period is obtained.
  • the vehicle model demand forecast data includes the forecast demand of multiple pre-selected vehicle models; the product selection module is specifically used for: according to the predicted demand of each pre-selected vehicle model in the multiple pre-selected vehicle models, through demand transformation Calculate the first actual demand of each central warehouse in the target city for each pre-selected vehicle model within a preset time period; for each central warehouse, the first actual demand corresponding to each pre-selected vehicle model Sorting is performed according to the order from high to low; all pre-selected car models in the preset ranking order are used as the product selection results of the central warehouse.
  • the demand conversion rate is obtained by comparing the historical demand with the predicted value corresponding to the historical demand; the predicted value corresponding to the historical demand is obtained during the training process of the decision tree model .
  • the device further includes: a first adjustment module 2305; a first adjustment module configured to, after determining the product selection result of each of the central warehouses, push the vehicle model corresponding to the product selection result to each The user terminal, so that users of each user terminal provide feedback on the vehicle model corresponding to the product selection result; receive feedback information from each user terminal, and adjust the product selection result according to the feedback information.
  • a first adjustment module 2305 configured to, after determining the product selection result of each of the central warehouses, push the vehicle model corresponding to the product selection result to each The user terminal, so that users of each user terminal provide feedback on the vehicle model corresponding to the product selection result; receive feedback information from each user terminal, and adjust the product selection result according to the feedback information.
  • the device further includes: a first to-be-added data determination module 2306; a first to-be-added data determination module 2306, configured to determine the product selection results of each of the central warehouses according to the The product selection result of the center warehouse and the current number of each vehicle type in each center warehouse are calculated, and the number of each vehicle type to be supplemented in each center warehouse is calculated.
  • a first to-be-added data determination module 2306 configured to determine the product selection results of each of the central warehouses according to the The product selection result of the center warehouse and the current number of each vehicle type in each center warehouse are calculated, and the number of each vehicle type to be supplemented in each center warehouse is calculated.
  • the device further includes: a second to-be-added data determination module 2307; a second to-be-added data determination module 2307, configured to, after the adjustment of the product selection result, correspond to each of the central warehouses Calculate the number of vehicle types to be added for each central warehouse based on the adjusted product selection results of each of the central warehouses and the current number of vehicle types in each of the central warehouses.
  • a second to-be-added data determination module 2307 configured to, after the adjustment of the product selection result, correspond to each of the central warehouses Calculate the number of vehicle types to be added for each central warehouse based on the adjusted product selection results of each of the central warehouses and the current number of vehicle types in each of the central warehouses.
  • the training samples are stored in a sample library; the device further includes: a first update module 2108; an update module 2308, configured to obtain the product selection result of each of the central warehouses after the determination The second actual demand of each central warehouse in the target city for vehicle model demand within a preset time period; update the sample library according to the characteristic data and the second actual demand.
  • the device further includes: a second adjustment module 2309; a second adjustment module 2309, configured to, after determining the product selection result of each of the central warehouses, obtain the pre-determined status of each central warehouse in the target city.
  • a second adjustment module 2309 configured to, after determining the product selection result of each of the central warehouses, obtain the pre-determined status of each central warehouse in the target city.
  • the vehicle deployment device of the embodiment shown in FIG. 23 can be used to implement the technical solutions of the foregoing method embodiments, and its implementation principles and technical effects are similar, and will not be repeated here.
  • the foregoing embodiment obtains characteristic data of a target city in a first preset historical time period, where the target city is provided with at least one central warehouse, and the characteristic data is used to represent the environmental information of each central warehouse in the target city And the associated data of the vehicle types in the target city; then according to the feature data, through the prediction model trained by the decision tree model, the vehicle demand forecast data corresponding to each central warehouse in the target city within the preset time period is obtained; and then; According to the vehicle demand forecast data, the product selection results of each of the central warehouses are determined.
  • Fig. 24 is an exemplary structure diagram of a vehicle deployment device 2400 according to some embodiments of the present application.
  • the vehicle deployment equipment may specifically be the car rental platform in the foregoing embodiment.
  • the vehicle deployment device can execute the processing flow provided by the method embodiment of the vehicle deployment device.
  • the device 2400 provided in this embodiment includes: at least one processor 2401 and a memory 2402. Among them, the processor 2401 and the memory 2402 are connected through a bus 2403.
  • At least one processor 2401 executes the computer-executable instructions stored in the memory 2402, so that at least one processor 2401 executes the method in the foregoing method embodiment.
  • the processor may be a central processing unit (English: Central Processing Unit, abbreviated as: CPU), and may also be other general-purpose processors or digital signal processors (English: Digital Signal Processor, referred to as DSP), application specific integrated circuit (English: Application Specific Integrated Circuit, referred to as ASIC), etc.
  • the general-purpose processor may be a microprocessor or the processor may also be any conventional processor or the like. The steps of the method disclosed in combination with the invention can be directly embodied as executed and completed by a hardware processor, or executed and completed by a combination of hardware and software modules in the processor.
  • the memory may include a high-speed RAM memory, and may also include a non-volatile storage NVM, such as at least one disk memory.
  • NVM non-volatile storage
  • the bus can be an Industry Standard Architecture (ISA) bus, Peripheral Component Interconnect (PCI) bus, or Extended Industry Standard Architecture (EISA) bus, etc.
  • ISA Industry Standard Architecture
  • PCI Peripheral Component Interconnect
  • EISA Extended Industry Standard Architecture
  • the bus can be divided into address bus, data bus, control bus and so on.
  • the buses in the drawings of this application are not limited to only one bus or one type of bus.
  • a computer-readable storage medium is also provided, on which a computer program is stored, and the computer program is executed by a processor to implement the vehicle deployment method described in the foregoing embodiment.
  • FIG. 25 is an exemplary flowchart of a vehicle deployment method 2500 according to some embodiments of the present application.
  • the method 2500 may be executed by a processor.
  • Step 2501 Obtain the associated feature data of the target car rental location.
  • the associated data features include: historical turnover vehicle information and related information of the target car rental location and storage location information of the target car rental area.
  • Step 2502 based on the prediction model, process the historical turnover vehicle information and related information of the target car rental location and the storage location information of the target car rental area, and determine the number of vehicles required at the target car rental location.
  • Step 2503 Determine a vehicle scheduling plan for the target car rental spot based on at least the number of vehicles required by the target car rental spot.
  • the historical turnover vehicle information and related information of the target car rental location and the location information of the target car rental area are processed to reduce the shortcomings of timeliness, unreasonable, inflexible, low efficiency, high labor cost, etc. Efficiently predict and determine the demand for the number of vehicles to be deployed at the target car rental location in the target car rental area, so as to achieve a more flexible, reasonable and effective vehicle deployment plan, and ultimately improve the efficiency of car rental business, meet user needs, and improve user experience.
  • FIG. 26 is an exemplary flowchart of a vehicle deployment method 2600 according to some embodiments of the present application.
  • the method 2600 may be executed by a processor.
  • Step 2601 Obtain the associated feature data of the target car rental location.
  • the associated data features include: historical turnover vehicle information and related information of the target car rental location and storage location information of the target car rental area.
  • Step 2602 based on the prediction model, process the historical turnover vehicle information and related information of the target car rental location and the storage location information of the target car rental area, and determine the number of vehicles required at the target car rental location.
  • Step 2603 Based on the third attribute information of the vehicles to be dispatched and the number of vehicles required by the target car rental point, the dispatched vehicles of the vehicle dispatch plan are determined from the vehicles to be dispatched.
  • FIG. 27 is an exemplary flowchart of a vehicle deployment method 2700 according to some embodiments of the present application.
  • the method 2700 may be executed by a processor.
  • Step 2701 Obtain the associated feature data of the target car rental location.
  • the associated data features include: the historical turnover of the target car rental location and the multi-dimensional vehicle turnover impact data and the number of storage gaps in the target car rental area.
  • Step 2702 Based on the prediction model, process the historical turnover number of vehicles in the target car rental location, multi-dimensional vehicle turnover impact data, and the number of gaps in the target car rental area, and determine the target car rental location corresponding to the vehicle number quota demand in the target car rental area.
  • Step 2703 Based on the third vehicle attribute information of the vehicles to be dispatched and the vehicle number quota requirement of the target car rental point, the dispatched vehicles of the vehicle dispatch plan are determined from the vehicles to be dispatched.
  • the associated feature data selects the historical turnover vehicle number and multi-dimensional vehicle turnover impact data of the target car rental point, and collects data reflecting the vehicle turnover situation of the target car rental point more comprehensively from multiple angles, so that the prediction and calculation through the predictive model can be used to obtain a more suitable target
  • the actual allocation of vehicle demand at the rental point and then determine the dispatched vehicles from the vehicles to be dispatched according to the vehicle number quota requirements of the target rental point, which improves the efficiency of vehicle allocation and reduces the operating cost of manual statistical allocation, thereby promoting the efficient operation of the rental business and meeting More timely and flexible needs of users improve user experience.
  • the associated feature data of the target car rental point may select the historical turnover vehicle number of the target car rental point and the multi-dimensional vehicle turnover impact data and the number of storage gaps in the target car rental area, and the third vehicle attribute information of the vehicle to be dispatched may be Select the preset vehicle attribute information of the vehicles to be dispatched, and determine the vehicle dispatch plan of the target car rental point through predictive model processing.
  • FIG. 28 an exemplary flowchart of a vehicle deployment method 2800 provided by some embodiments of the present application is shown.
  • the method 2800 may be implemented by a processor.
  • Step 2801 Obtain the number of gapped vehicles in each central warehouse of the leasing company as the initial storage quota of each central warehouse;
  • Step 2802 For each central warehouse, obtain the true delivery peak value of the central warehouse in the historical time period, calculate the predicted delivery peak value of the central warehouse in the same historical time period according to the initial warehouse location quota, and calculate according to the actual delivery peak value and For the difference between the predicted peak delivery value, the initial location quota is corrected through a prediction model to obtain the corrected location quota of the central warehouse;
  • Step 2803 Allocate the corresponding vehicles of the revised storage space quota to each central warehouse of the leasing company.
  • the vehicle allocation method is mainly applied to the server of the car rental platform, and the server makes the allocation decision for each rental company.
  • a leasing company generally has multiple central warehouses, and step 2801 obtains the number of vehicles in each central warehouse that the leasing company has gaps in. The number of vehicles in the gap can be determined through prediction, or through sensor monitoring, or through staff uploading data. The number of vehicles in the gap is used to indicate how many locations need to be added or reduced for the leasing company.
  • step 2802 corrects the initial location quota of each central warehouse, so that the location quota is more in line with the real demand.
  • step 2803 allocates the corresponding vehicles of the revised storage space quota to each central warehouse of the leasing company.
  • the allocation method can be realized by way of releasing a work order.
  • the vehicle allocation method realizes the intelligent management of the warehouse locations of the car rental business center warehouse, can dynamically allocate warehouse locations for each rental company, and supports the increase in the number of center warehouses and the number of leases, and the change of the total warehouse locations of each center warehouse , And then expand to other cities on a large scale, as long as you provide some basic information and order data of the warehouse, you can quickly expand to other cities.
  • this method greatly saves operating manpower, and improves the utilization rate of parking spaces and vehicle turnover rate, and fully meets the needs of leasing. It will not happen that the leasing fails to meet the user's ordering needs due to unfair or insufficient allocation of storage spaces. , Making the cooperation between the platform and leasing more efficient.
  • the location quota is revised to make the location quota more in line with the real demand.
  • FIG. 29 an exemplary flowchart of a vehicle deployment method 2900 provided by some embodiments of the present application is shown.
  • the method 2900 may be implemented by a processor.
  • Step 2901 Obtain the historical rental car quantity of the rental company, the information dimension data of the rental company, and time data, input the prediction model, and obtain the total rental car quantity of the rental company output by the prediction model;
  • Step 2902 Obtain the historical number of cars returned by the leasing company, the information dimension data of the leasing company, and time data, input the prediction model, and obtain the total number of cars returned by the leasing company output by the prediction model;
  • Step 2903 Obtain the historical proportion of each central warehouse of the leasing company, determine that the number of rented cars in each central warehouse is the total number of rented cars ⁇ the historical proportion of the central warehouse, and the number of cars returned in each central warehouse is the total number of cars returned ⁇ The historical proportion of the central warehouse;
  • Step 2904 obtain the parking quantity of each central warehouse
  • Step 2905 based on the number of rented cars, the number of cars returned, and the number of parking in each central warehouse, determine the number of gapped vehicles in each central warehouse as the initial storage quota for each central warehouse;
  • Step 2907 If the difference between the predicted peak delivery value minus the actual delivery peak value is greater than the difference threshold, use the initial location quota as the revised location quota;
  • step 2908 if the difference between the predicted delivery peak minus the real delivery peak is less than or equal to the difference threshold, calculate the revised location quota as: (real delivery peak + difference threshold)/(N/A);
  • Step 2909 Allocate the corresponding vehicles of the revised warehouse location quota to each central warehouse of the leasing company.
  • step 2901 and step 2902 use the prediction model to predict the total number of rental cars and the total number of cars returned by the leasing company.
  • the prediction model Preferably, by inputting historical car rental order volume, rental company scale data, holiday information, weather, rental company credit score, car model richness, rental brand awareness and other dimensional features into the prediction model, the regression of the rental car order volume is performed, and each The forecast data of the rental company in the next X hours.
  • X is a variable parameter.
  • the prediction of the number of cars returned is the same as above.
  • the data is processed and input into the prediction model to predict the number of cars returned, through the model
  • the forecast outputs the forecast data of the number of cars returned by each leasing company in the next X hours.
  • step 2903 the total number of rented cars and the total number of returned cars are determined according to the historical proportion of each central warehouse through the historical proportions to determine the number of rented cars and the number of returned cars in each central warehouse.
  • step 2904 the number of gapped vehicles in each central warehouse of each leasing company can be calculated.
  • step 2905 can calculate how many more locations need to be added or reduced for each central warehouse of each leasing company as the initial location quota of the central warehouse.
  • step 2906 to step 2908 will correct the initial location quota through real-time data.
  • data correction is carried out based on the actual number of vehicles delivered in the last N days.
  • the reason for dividing by A in the formula is related to the business background. It takes a certain amount of time for a car to leave the warehouse and enter the warehouse. Normally, a car turns over a car within the parking space flow rate A.
  • the purpose of this module is to ensure that it is based on the model. The predicted number of allocated warehouses is sufficient to meet the actual vehicle turnover rate of each leasing company.
  • the total number of rented cars and the total number of returned cars of the leasing company are predicted by the prediction model, and then the total number of rented cars and the total number of returned cars are allocated to each central warehouse based on the historical proportion of each central warehouse. Since the data of the entire leasing company is used for forecasting, there is a large enough data sample to make the overall data accurate, and the distribution of historical proportions can also accurately reflect the situation of each central warehouse. Finally, through the correction of the storage space quotas, large errors in the prediction were avoided. At the same time, the flow rate of parking spaces was introduced to predict the model, which was in line with the real vehicle turnover rate of each leasing company.
  • Fig. 30 is a schematic diagram of an exemplary system implemented by a vehicle deployment method according to some embodiments of the present application.
  • the acquisition module 501 can be used to acquire data from various parties, and perform data cleaning, feature processing, and structured storage.
  • the acquiring module 501 may include a storage location acquiring device 3011, a rental company data acquiring device 3012, a historical order quantity acquiring device 3013, and a weather and holiday data acquiring device 3014.
  • the first determining module 502 can be used to predict the demand for car rental and the number of cars returned through the prediction model, and provide decision-making data support for subsequent modules.
  • the first determining module 50 may include a rental car demand prediction module 3021, and a car return quantity prediction module 3022.
  • the second determination module 503 makes automated real-time decisions based on the predicted data and business scenarios.
  • the second determining module 503 may include a location quota decision module 3031 and a location quota correction module 3032.
  • the first layer is the data acquisition layer 3010, which is mainly responsible for acquiring data from all parties, and doing data cleaning, feature processing and structured storage, including position acquisition device 3011, rental company data acquisition device 3012, history Single-quantity acquisition device 3013, weather and holiday data acquisition device 3014;
  • the second layer is the basic model prediction layer 3020, which predicts rental car demand and the number of cars returned through the prediction model, and provides decision-making data support for subsequent modules, including car rental demand prediction Module 3021, and return quantity prediction module 3022;
  • the third layer is the business decision-making layer 3030, which makes automated real-time decisions based on predicted data and business scenarios, including the location quota decision module 3031 and the location quota correction module 3032 . among them:
  • Car rental demand forecasting module 3021 By inputting historical car rental orders, CP company scale data, holiday information, weather, CP company credit scores, car model richness, CP brand awareness and other dimensional features into the prediction model, the regression of car rental orders, Output the forecast data of each CP company renting out vehicles in the next X hours. (X hours can be flexibly changed as a parameter)
  • Return car quantity prediction module 3022 Same as above, through historical old user’s unit data returned after the rental period expires, return some characteristics and context information of CP company, and input the data into the prediction model to predict the number of cars returned. , Through the model forecast, output the forecast data of the number of cars returned by each CP company in the next X hours.
  • Location quota decision module 3031 Based on the predicted car rental demand data and the number of cars returned, we can know the cars to be rented out and the cars to be recovered by each CP company, and then this data is allocated to each car according to the historical proportion. Central warehouse, you can know the number of rented cars and the number of cars returned by each CP company to each central warehouse, then combined with the number of vehicles parked by each CP company in the central warehouse, you can calculate the number of each CP company The number of gapped vehicles. At this point, it can be calculated how many more locations need to be added or reduced for each CP company.
  • Location quota correction module 3032 When the business scale is small, if the historical data level is small, the value error predicted by the model will be relatively large. At this time, real-time data can be used to correct it. In order to ensure that the number of warehouses configured for each CP can meet its vehicle delivery in the next N days, we use the actual number of vehicles delivered in the last N days to make data corrections.
  • the revised calculation formula is: (CP real delivery peak + threshold)/(N/2) rounded up to get the revised location quota number. The reason for dividing by 2 in the formula is related to the business background.
  • This module is to ensure the prediction based on the model.
  • the number of allocated warehouses is enough for each CP company's true vehicle turnover rate.
  • the intelligent management of the warehouse locations of the car rental business center is realized, which can dynamically allocate warehouse locations for each CP company every day, and support the increase in the number of central warehouses and the number of CPs, as well as the total warehouse location of each central warehouse. Changes, and then large-scale expansion to other cities, as long as some basic information and order data of the warehouse can be quickly expanded to other cities.
  • this method greatly saves operating manpower, and improves the utilization rate of parking spaces and vehicle turnover rate, and fully meets the needs of CP. It will not happen that CP fails to meet the user's car booking needs due to unfair or insufficient storage space allocation. , Which makes the cooperation between the platform and CP more efficient.
  • the device 3100 includes:
  • At least one processor 3101 and,
  • the memory 3102 stores instructions executable by the one processor, and the instructions are executed by the at least one processor, so that the at least one processor can:
  • each central warehouse For each central warehouse, obtain the true delivery peak value of the central warehouse in the historical time period, calculate the predicted delivery peak value of the central warehouse in the same historical time period according to the initial warehouse location quota, and calculate according to the actual delivery peak value and the predicted peak value.
  • the difference of the delivery peak value is corrected by the prediction model to the initial location quota to obtain the corrected location quota of the central warehouse;
  • a processor 3101 is taken as an example.
  • the electronic device is preferably a server of the car rental platform.
  • the electronic device may further include: an input device 3103 and a display device 3104.
  • the processor 3101, the memory 3102, the input device 3103, and the display device 3104 may be connected by a bus or other methods, and the connection by a bus is taken as an example in the figure.
  • the memory 3102 can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules, such as the program instructions corresponding to the vehicle deployment method in the embodiments of the present application /Module, for example, the method flow shown in Figure 28.
  • the processor 3101 executes various functional applications and data processing by running the non-volatile software programs, instructions, and modules stored in the memory 3102, that is, realizes the parking space allocation method of the central warehouse of the car rental platform in the foregoing embodiment.
  • the memory 3102 may include a storage program area and a storage data area, where the storage program area can store an operating system and an application program required by at least one function; the storage data area can store data created according to the use of the parking space allocation method of the central warehouse of the car rental platform Wait.
  • the memory 3102 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage devices.
  • the memory 3102 may optionally include memories remotely provided with respect to the processor 3101, and these remote memories may be connected to a device that executes the method for allocating parking spaces in the central warehouse of the car rental platform through a network. Examples of the aforementioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.
  • the input device 3103 can receive inputted user clicks, and generate signal inputs related to user settings and function control of the parking space allocation method of the central warehouse of the car rental platform.
  • the display device 3104 may include a display device such as a display screen.
  • the one or more modules are stored in the memory 3102, and when run by the one or more processors 3101, the method for allocating parking spaces in the central warehouse of the car rental platform in any of the foregoing method embodiments is executed.
  • This embodiment realizes the intelligent management of the warehouse locations of the car rental business center warehouse, can dynamically allocate warehouse locations for each rental company, and supports the increase in the number of center warehouses and the number of leases, and the change of the total warehouse locations of each center warehouse. And then it can be expanded to other cities on a large scale, as long as some basic information and order data of the warehouse are provided, it can be quickly expanded to other cities.
  • this method greatly saves operating manpower, and improves the utilization rate of parking spaces and vehicle turnover rate, and fully meets the needs of leasing. It will not happen that the leasing fails to meet the user's ordering needs due to unfair or insufficient allocation of storage spaces. , Making the cooperation between the platform and leasing more efficient.
  • the present invention corrects the location quota based on the actual delivery peak value, so that the location quota is more in line with the real demand.
  • Some embodiments of the present application provide an electronic device for vehicle deployment, including:
  • At least one processor and,
  • a memory communicatively connected with the at least one processor; wherein,
  • the memory stores instructions executable by the one processor, and the instructions are executed by the at least one processor, so that the at least one processor can:
  • each central warehouse of the leasing company determines that the number of rented cars in each central warehouse is the total number of rented cars ⁇ the historical proportion of the central warehouse, and the number of cars returned in each central warehouse is the total number of returned cars ⁇ the central warehouse % Of history;
  • the number of gapped vehicles in each central warehouse is determined as the initial storage space quota for each central warehouse
  • the initial location quota is used as the revised location quota
  • the revised location quota is calculated as: (real delivery peak + difference threshold)/(N/A);
  • the total number of rented cars and the total number of returned cars of the leasing company are predicted by the prediction model, and then the total number of rented cars and the total number of returned cars are allocated to each central warehouse based on the historical proportion of each central warehouse. Since the data of the entire leasing company is used for forecasting, there is a large enough data sample to make the overall data accurate, and the distribution of historical proportions can also accurately reflect the situation of each central warehouse. Finally, through the correction of the storage space quotas, large errors in the prediction were avoided. At the same time, the flow rate of parking spaces was introduced to predict the model, which was in line with the real vehicle turnover rate of each leasing company.
  • the processing device may obtain the historical rental car number, weather data, and historical return number of the rental company to form a combined feature sequence, and input the combined feature sequence into an embedded model based on the RNN model to obtain a vehicle flow representation vector. Further, the characteristics of the vehicle flow representation vector and the information dimension data of the leasing company are input into the prediction model to obtain the total number of rental cars of the leasing company output by the prediction model.
  • the embedded model and the input prediction model can be obtained through joint training.
  • the combined feature sequence is composed of feature combination values at several time points.
  • the feature combination value at each time point is formed by the combination of the number of historical car rentals, weather data, and historical car return data at that time point, and is multiplied by the time weight coefficient.
  • the time weight coefficient can be different depending on the distance of the time point, and the weight coefficient of the time point closer to the current time can be larger.
  • the historical value of the number of cars returned can be prescribed to reduce the impact of this value in the prediction of car rental.
  • the exponent of the square root may be between 0.4 and 0.6, for example, the 0.5 power of the historical number of cars returned is used to calculate the feature combination value.
  • the influence of different factors on the forecast results can be better reflected, especially the interrelationship between these factors.
  • the impact of weather is related to the front and back
  • the impact of the number of car rentals and returns is also related.
  • RNN-based processing can reflect the relationship between the front and back time points and make the prediction results more accurate.
  • the embodiment of the present application also provides a storage medium for storing computer instructions, and when the computer executes the computer instructions, it is used to execute all the steps of the aforementioned vehicle deployment method.
  • the vehicle deployment and its hardware device can be applied to the following vehicle deployment scenarios.
  • the back-end server of the car rental platform obtains the rental company’s historical turnover number of vehicles and multi-dimensional vehicle turnover impact data from the rental company’s server or other data sources; the back-end server of the rental car platform sends the central warehouse’s location gap to the central warehouse server In response to the number of data acquisition requests, the central warehouse server sends corresponding data to the back-end server of the car rental platform; the server of the car rental platform transmits the corresponding data to the back-end server of the car rental platform according to the historical turnover of vehicles and multi-dimensional vehicle turnover impact data and the number of storage gaps.
  • the data is processed through the predictive model, and the vehicle scheduling plan for vehicle deployment is finally determined.
  • the data is updated in real time and sent to the server of the rental company and the server of the central warehouse, whether it is a lease
  • the company's supplier users and the management users of the central warehouse can log in to the corresponding client of the car rental platform through their terminals to obtain and query vehicle deployment information data (such as vehicle deployment model information, deployment progress information, etc.).
  • the associated feature sample data of the target car rental point and the associated vehicle demand label data can be used to train the model to be trained.
  • the model to be trained can use artificial intelligence algorithms, specifically, decision trees, random forests, logistic regression, support vector machines, naive Bayes, K-nearest neighbor algorithm, K-means algorithm, Adaboost (a kind of Boosting algorithm) , Neural Network, Markov's Machine Learning Algorithm.
  • the associated feature sample data is the input of the model, and the associated vehicle demand label data corresponding to the associated feature sample data (for example, the label data created from the historical associated vehicle demand of the associated feature sample data) is used as the label, and the output of the model is the associated vehicle demand forecast Data, training the model to be trained.
  • the associated vehicle demand label data corresponding to the associated feature sample data for example, the label data created from the historical associated vehicle demand of the associated feature sample data
  • the output of the model is the associated vehicle demand forecast Data, training the model to be trained.
  • the output can be compared with the label to calculate the error, and then the model parameters can be fed back and adjusted until the error between the output and the label is small and stable, that is, the model achieves the desired training effect , Use the trained model as a predictive model.
  • the training process 3200 may be executed by a processor.
  • FIG. 32 is an exemplary flowchart of a prediction model training process 3200 according to some embodiments of the present application.
  • the prediction model can be trained through the following steps:
  • Step 3201 Obtain the associated feature sample data of the target car rental point and its associated vehicle demand label data
  • Step 3202 input the associated feature sample data into the model to be trained, and output the prediction result data;
  • Step 3203 Update the model parameters according to the prediction result data and the associated vehicle demand tag data, and continue training until the prediction model is obtained.
  • the feature quantities of multiple dimensions corresponding to each central warehouse of each predetermined city can be selected as the associated feature sample data, and the historical demand corresponding to each central warehouse of each predetermined city can be used as the associated vehicle demand label data.
  • the prediction model is obtained through the decision tree model training. For more details and related descriptions of the prediction model training, please refer to Figure 19 and Figure 20, which will not be repeated here.
  • the target car rental point can be uploaded to the server, the server of the car rental platform, and/or the business backend server that has a data communication connection relationship with the target car rental point server and the target car rental area server to obtain the target car rental point. Associate feature sample data and its associated vehicle demand label data.
  • the prediction model with the expected prediction effect can be obtained, so that the prediction model can compare the associated feature data of a variety of target car rental points.
  • the processing process is more efficient, the prediction accuracy and adaptability are higher, and it is closer to the vehicle demand information obtained in the user's vehicle deployment business.
  • a person of ordinary skill in the art can understand that all or part of the steps in the foregoing method embodiments can be implemented by a program instructing relevant hardware.
  • the aforementioned program can be stored in a computer readable storage medium. When the program is executed, it executes the steps including the foregoing method embodiments; and the foregoing storage medium includes: ROM, RAM, magnetic disk, or optical disk and other media that can store program codes.
  • this application uses specific words to describe the embodiments of the application.
  • “one embodiment”, “an embodiment”, and/or “some embodiments” mean a certain feature, structure, or characteristic related to at least one embodiment of the present application. Therefore, it should be emphasized and noted that “one embodiment” or “one embodiment” or “an alternative embodiment” mentioned twice or more in different positions in this application does not necessarily refer to the same embodiment. .
  • some features, structures, or characteristics in one or more embodiments of the present application can be appropriately combined.
  • numbers describing the number of ingredients and attributes are used. It should be understood that such numbers used in the description of the embodiments use the modifier "about”, “approximately” or “substantially” in some examples. Retouch. Unless otherwise stated, “approximately”, “approximately” or “substantially” indicates that the number is allowed to vary by ⁇ 20%.
  • the numerical parameters used in the specification and claims are approximate values, and the approximate values can be changed according to the required characteristics of individual embodiments. In some embodiments, the numerical parameter should consider the prescribed effective digits and adopt the method of general digit retention. Although the numerical ranges and parameters used to confirm the breadth of the range in some embodiments of the present application are approximate values, in specific embodiments, the setting of such numerical values is as accurate as possible within the feasible range.

Abstract

一种车辆调配方法、装置、设备及计算机可读存储介质，所述车辆调配方法，包括通过处理器执行以下步骤：获取目标租车点的关联特征数据，所述关联特征数据包括所述目标租车点和目标租车区域中的至少一种的相关特征数据；所述目标租车区域为与所述目标租车点相关的区域（410）；基于预测模型对所述目标租车点的关联特征数据进行处理，确定所述目标租车点的关联车辆需求，所述关联车辆需求为所述目标租车点在所述目标租车区域中对车辆的相关需求（420）；至少基于所述关联车辆需求，确定所述目标租车点的车辆调度方案（430）。通过利用预测模型处理目标租车点影响车辆调配业务的关联特征数据，提供了一种调配更合理、满足多种需求的高效车辆调配方案。

Description

一种车辆调配方法、装置、设备及计算机可读存储介质
交叉引用
本申请要求2019年12月26日递交的申请号为201911369388.1的中国申请的优先权，以及2020年1月3日递交的申请号为202010006858.4的中国申请的优先权和2020年3月12日递交的申请号为202010171414.6的中国申请的优先权，其所有内容通过引用的方式包含于此。
技术领域
本申请涉及数据处理技术领域，特别涉及一种车辆调配方法、装置、设备及计算机可读存储介质。
背景技术
在租车业务场景中，一般是租车平台根据业务需求确定车型等车辆信息，然后再通过租赁公司等车辆供应商竞价的方式，选购相应车辆，之后通过多道验车工序对选购的车辆进行检验，检验之后存放在每个目标城市的相应中心仓。用户通过线下租车门店或线上应用选购下单需要租用的车辆后，根据指定看车或提车地点完成租车程序。显然，无论是在租车业务链上的哪道流程，都涉及到车辆供应商、中心仓、租车门店、租赁公司等根据车辆业务需求的车辆调配问题。在进行车辆调配时，需要在满足相应车型品类、需求数量等基础指标的同时，还需要兼顾在多方之间进行车辆调配的整体效率甚至是细节需求。
目前，传统车辆调配业务场景中存在车辆调配效率低下、车辆分配不够灵活合理、效率低的技术现状。例如，用户需要前往中心仓看车并提车，而由于中心仓在城市分布密度较低，传统的集中中心仓车辆分配方式，不能满足用户就近看车提车的需求，降低了车辆调配效率，进而影响租车效率；通过传统的平台网约车型数据信息进行简单排序及人工筛选来确定车辆调配的车型选品需求，不 仅效率低，严重耗费运营人力成本，还由于没有兼顾市场变动、地域环境等特点、车辆供应商动态等现实因素，导致车辆调配不能及时有效地贴合实际调配需求以满足市场租车需求，从而影响租车效率；在每个目标城市中有多个中心仓，而每个中心仓中又有多个库位，如何通过高效的车辆调配，保障这些库位的高效使用，以及满足多个合作租赁公司(CP)的需求是租车业务中非常重要的一环，现有库位分配的整体工作流是偏线下、人工、滞后的；中心仓车位的管理往往是仅通过人工统计租赁公司占有率排序的简单业务逻辑确定租赁公司库位配额，存在缺乏时效性、库位分配数量不合理、不灵活、效率低、人工成本高等多重缺陷，不利于车辆调配的合理有效进行以及进一步的效率提高，增加了运营负担，从而也影响整体的租车效率。因此，亟需提供一种调配更合理、满足多种租车业务场景需求的高效车辆调配方案。
发明内容
本申请实施例之一提供一种车辆调配方法。所述车辆调配方法包括：获取目标租车点的关联特征数据，所述关联特征数据包括所述目标租车点和目标租车区域中的至少一种的相关特征数据；所述目标租车区域为与所述目标租车点相关的区域；基于预测模型对所述目标租车点的关联特征数据进行处理，确定所述目标租车点的关联车辆需求，所述关联车辆需求为所述目标租车点在所述目标租车区域中对车辆的相关需求；至少基于所述关联车辆需求，确定所述目标租车点的车辆调度方案。
本申请实施例之一提供一种车辆调配装置，包括处理器，所述处理器包括：获取模块，用于：获取目标租车点的关联特征数据，所述关联特征数据包括所述目标租车点和目标租车区域中的至少一种的相关特征数据；所述目标租车区域为与所述目标租车点相关的区域；第一确定模块，用于：基于预测模型对所述目标租车点的关联特征数据进行处理，确定所述目标租车点的关联车辆需求，所述关联车辆需求为所述目标租车点在所述目标租车区域中对车辆的相关 需求；第二确定模块，用于：至少基于所述关联车辆需求，确定所述目标租车点的车辆调度方案。
本申请实施例之一提供一种车辆调配设备，包括：存储器；处理器；以及计算机程序；其中，所述计算机程序存储在所述存储器中，并被配置为由所述处理器执行以实现如上述所述的方法。
本申请实施例之一提供一种计算机可读存储介质，其上存储有计算机程序，所述计算机程序被处理器执行时实现如上述所述的方法。
附图说明
本申请将以示例性实施例的方式进一步说明，这些示例性实施例将通过附图进行详细描述。这些实施例并非限制性的，在这些实施例中，相同的编号表示相同的结构，其中：
图1是根据本申请一些实施例所示的车辆调配系统100的应用场景示意图；
图2是根据本申请一些实施例所示的示例性计算设备200的示例性硬件和/或软件的示意图；
图3是根据本申请一些实施例所示的示例性移动设备300的示例性硬件和/或软件的示意图；
图4是根据本申请一些实施例所示的车辆调配方法400的示例性流程图；
图5是根据本申请一些实施例所示的车辆调配装置500的示例性结构图；
图6是根据本申请一些实施例所示的车辆调配设备600的示例性结构图；
图7是根据本申请一些实施例所示的车辆调配方法700的示例性流程图；
图8是根据本申请一些实施例所示的车辆调配方法800的示例性流程图；
图9是根据本申请一些实施例所示的车辆调配方法900的示例性流程图；
图10是根据本申请一些实施例所示的车辆调配方法1000的示例性流程图；
图11是根据本申请一些实施例所示的车辆调配方法1100的示例性流程图；
图12是根据本申请一些实施例所示的车辆调配装置1200的示例性结构图；
图13是根据本申请一些实施例所示的用于车辆调配的电子设备1300的示例性结构图；
图14是根据本申请一些实施例所示的车辆调配方法1400的示例性流程图；
图15是根据本申请一些实施例所示的车辆调配方法1500的示例性流程图；
图16是根据本申请一些实施例所示的车辆调配方法1600的示例性流程图；
图17是根据本申请一些实施例所示的车辆调配方法及其硬件装置应用于用户使用租车平台的应用场景示意图；
图18是根据本申请一些实施例所示的车辆调配方法1800的示例性流程图；
图19是根据本申请一些实施例所示的车辆调配方法1900的示例性流程图；
图20是根据本申请一些实施例所示的车辆调配方法2000的流程示意图；
图21是根据本申请一些实施例所示的车辆调配方法2100的示例性流程图；
图22是根据本申请一些实施例所示的车辆调配方法2100的应用场景示例图；
图23为根据本申请一些实施例所示的车辆调配装置2300的示例性结构图；
图24是根据本申请一些实施例所示的车辆调配设备2400的示例性结构 图；
图25是根据本申请一些实施例所示的车辆调配方法2500的示例性流程图；
图26是根据本申请一些实施例所示的车辆调配方法2600的示例性流程图；
图27是根据本申请一些实施例所示的车辆调配方法2700的示例性流程图；
图28是根据本申请一些实施例提供的车辆调配方法2800的示例性流程图；
图29是根据本申请一些实施例提供的车辆调配方法2900的示例性流程图；
图30是根据本申请一些实施例提供的车辆调配方法实现的示例性系统原理图；
图31是根据本申请一些实施例提供的车辆调配设备3100的示例性结构图；
图32是根据本申请一些实施例所示的预测模型训练过程3200的示例性流程图。
具体实施方式
为了更清楚地说明本申请实施例的技术方案，下面将对实施例描述中所需要使用的附图作简单的介绍。显而易见地，下面描述中的附图仅仅是本申请的一些示例或实施例，对于本领域的普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图将本申请应用于其它类似情景。除非从语言环境中显而易见或另做说明，图中相同标号代表相同结构或操作。
应当理解，本文使用的“系统”、“装置”、“单元”和/或“模块”是用于区分不同级别的不同组件、元件、部件、部分或装配的一种方法。然而，如果其他词语 可实现相同的目的，则可通过其他表达来替换所述词语。
如本申请和权利要求书中所示，除非上下文明确提示例外情形，“一”、“一个”、“一种”和/或“该”等词并非特指单数，也可包括复数。一般说来，术语“包括”与“包含”仅提示包括已明确标识的步骤和元素，而这些步骤和元素不构成一个排它性的罗列，方法或者设备也可能包含其它的步骤或元素。
本申请中使用了流程图用来说明根据本申请的实施例的系统所执行的操作。应当理解的是，前面或后面操作不一定按照顺序来精确地执行。相反，可以按照倒序或同时处理各个步骤。同时，也可以将其他操作添加到这些过程中，或从这些过程移除某一步或数步操作。
图1是根据本申请的一些实施例所示的车辆调配系统100的应用场景示意图。
在应用场景中可以包括服务器110、网络120、数据查询端130、数据库140和其他数据源150。服务器110可包含处理设备112。
在一些实施例中，车辆调配系统100可以通过实施本申请中披露的方法和/或过程来进行车辆调配。在一些实施例中，服务器110可以布设在公共的租车平台，也可以布设在目标租车区域或目标租车点，服务器110可以用于经由网络120从数据库140或其他数据源150获取目标租车点的关联特征数据，并进行相应数据处理，获取针对目标租车点的车辆调度方案数据。数据库140和其他数据源150中的数据可以来自上传有包括目标租车点、目标租车区域产生的车辆调配业务相关数据的租车平台，也可以来自目标租车点或目标租车区域。数据内容包括目标租车点的关联特征数据、关联需求数据、车辆调度方案数据及其他车辆调配业务相关数据、目标租车区域的车辆调配业务相关数据、预测模型训练样本数据、预测模型结果数据等等。
在一些应用场景中，车辆调配系统100可以用于基于目标线下租车门店与目标中心仓的车辆调配关联特征数据、目标中心仓与目标线下租车门店的指定区域的车辆调配关联特征数据、目标租赁公司与其指定存放车辆的目标中心 仓的车辆调配关联特征数据，又或者目标中心仓与目标城市的车辆调配关联特征数据等多种车辆调配场景需求的数据处理，从而获取相应车辆调度方案。
示例性地，当车辆调配系统100用于涉及目标租赁公司与其指定存放车辆的目标中心仓的车辆调配具体场景时，可以通过服务器110从数据库140和其他数据源150获取目标租赁公司的关联特征数据(例如可以是目标租赁公司的历史车型数据、历史库位需求数据等)，然后通过执行程序指令进行借由预测模型的数据处理过程，最终确定目标租赁公司的车辆调度方案，以指导目标租赁公司如何向其指定存放车辆的目标中心仓调度车辆。
车辆调配业务期间产生的所有业务相关数据都可以存储在服务器110和/或数据库140和其他数据源150。预测模型及其相应数据可以直接存储在服务器110或其他存储设备(根据场景需要设定)中，也可以存储在数据库140或其他数据源150中。数据查询端130的用户可以包括中心仓、线下租车门店、租赁公司、租车平台、需要租车的用户或其他与租车业务或车辆调配相关的可能用户，至于数据查询权限可以根据具体应用场景不同进行相应设定。
服务器110与数据查询端130可以通过网络120相连，数据库140可以与服务器110通过网络120相连，也可以直接连接于服务器110或者处于服务器110的内部。数据库140、其他数据源150可与网络120连接以与车辆调配系统100的一个或多个组件通讯。车辆调配系统100的一个或多个组件可通过网络120访问存储于数据查询端130、数据库140和其他数据源150中的资料或指令。
在一些实施例中，服务器110、数据查询端端130以及其他可能的系统组成部分中可以包括数据库140。
在一些实施例中，服务器110、数据查询端端130以及其他可能的系统组成部分中可以包括处理器。
服务器110可以用于管理资源以及处理来自本系统至少一个组件或外部数据源(例如，云数据中心)的数据和/或信息。在一些实施例中，服务器110可 以是单一服务器或服务器组。该服务器组可以是集中式或分布式的(例如，服务器110可以是分布式系统)，可以是专用的也可以由其他设备或系统同时提供服务。在一些实施例中，服务器110可以是区域的或者远程的。在一些实施例中，服务器110可以在云平台上实施，或者以虚拟方式提供。仅作为示例，所述云平台可以包括私有云、公共云、混合云、社区云、分布云、内部云、多层云等或其任意组合。
在一些实施例中，服务器110可包含处理设备112。处理设备112可以采用相应处理器，处理从其他设备或系统组成部分中获得的数据和/或信息。处理器可以基于这些数据、信息和/或处理结果执行程序指令，以执行一个或多个本申请中描述的功能。在一些实施例中，处理设备112可以包含一个或多个子处理设备(例如，单核处理设备或多核多芯处理设备)。仅作为示例，处理设备112可以包括中央处理器(CPU)、专用集成电路(ASIC)、专用指令处理器(ASIP)、图形处理器(GPU)、物理处理器(PPU)、数字信号处理器(DSP)、现场可编程门阵列(FPGA)、可编辑逻辑电路(PLD)、控制器、微控制器单元、精简指令集电脑(RISC)、微处理器等或以上任意组合。
网络120可以连接系统的各组成部分和/或连接系统与外部资源部分。网络120使得各组成部分之间，以及与系统之外其他部分之间可以进行通讯，促进数据和/或信息的交换。在一些实施例中，网络120可以是有线网络或无线网络中的任意一种或多种。例如，网络120可以包括电缆网络、光纤网络、电信网络、互联网、局域网络(LAN)、广域网络(WAN)、无线局域网络(WLAN)、城域网(MAN)、公共交换电话网络(PSTN)、蓝牙网络、紫蜂网络(ZigBee)、近场通信(NFC)、设备内总线、设备内线路、线缆连接等或其任意组合。各部分之间的网络连接可以是采用上述一种方式，也可以是采取多种方式。在一些实施例中，网络可以是点对点的、共享的、中心式的等各种拓扑结构或者多种拓扑结构的组合。在一些实施例中，网络120可以包括一个或以上网络接入点。例如，网络120可以包括有线或无线网络接入点，例如基站和/或网络交换点120- 1、120-2、…，通过这些进出点系统100的一个或多个组件可连接到网络120上以交换数据和/或信息。
数据查询端130指用于数据查询的一个或多个终端设备或软件。在一些实施例中，使用数据查询端130的可以是一个或多个用户，可以包括直接使用服务的用户，也可以包括其他相关用户。在一些实施例中，数据查询端130可以是移动设备130-1、平板计算机130-2、膝上型计算机130-3、台式计算机等其他具有输入和/或输出功能的设备中的一种或其任意组合。
在一些实施例中，移动装置130-1可包括可穿戴设备和智慧移动装置等或其任意组合。
在一些实施例中，智能移动设备可以包括智能电话、个人数字助理(PDA)、游戏设备、导航设备、手持终端(POS)等或其任意组合。
在一些实施例中，台式计算机130-4可以是车载计算机、车载电视等。
例如，其他具有输入和/或输出功能的设备移动装置130-1可包括设置在公共场所的专用问答终端。
上述示例仅用于说明所述数据查询端130设备范围的广泛性而非对其范围的限制。
数据库140可以用于存储数据和/或指令。数据库140在单个中央服务器、通过通信链路连接的多个服务器或多个个人设备中实现。在一些实施例中，数据库140可包括大容量存储器、可移动存储器、挥发性读写存储器(例如，随机存取存储器RAM)、只读存储器(ROM)等或以上任意组合。示例性的，大容量储存器可以包括磁盘、光盘、固态磁盘等。在一些实施例中，数据库140可在云平台上实现。
其他数据源150可以用于为所述系统提供其他信息的一个或多个来源。其他数据源150可以是一个或多个设备，可以是一个或多个应用程序接口，可以是一个或多个数据库查询接口，可以是一个或多个基于协议的信息获取接口，可以是其他可获取信息的方式，可以是上述方式两种或多种的组合。信息源所提 供的信息，可以是在提取信息时已存在的，也可以是在提取信息时临时生成的，也可以是上述方式的组合。在一些实施例中，其他数据源150可以用于为系统提供环境信息、天气信息或其他与租车业务或车辆调配相关的任何可能信息。
图2是根据本申请一些实施例所示的示例性计算设备200的示例性硬件和/或软件的示意图。在一些实施例中，服务器110或数据查询端130可以在计算设备200上实现。例如，处理设备112可以在计算设备200上实施并执行本说明书所公开的处理设备112的功能。如图2所示，计算设备200可以包括总线210、处理器220、只读存储器230、随机存储器240、通信端口250、输入/输出260和硬盘270。处理器220可以执行计算指令(程序代码)并执行本说明书描述的车辆调配系统100的功能。计算指令可以包括程序、对象、组件、数据结构、过程、模块、功能(该功能指本说明书中描述的特定功能)等。例如，处理器220可以处理从车辆调配系统100的其他任何组件获取的图像或文本数据。在一些实施例中，处理器220可以包括微控制器、微处理器、精简指令集计算机(RISC)、专用集成电路(ASIC)、应用特定指令集处理器(ASIP)、中央处理器(CPU)、图形处理单元(GPU)、物理处理单元(PPU)、微控制器单元、数字信号处理器(DSP)、现场可编程门阵列(FPGA)、高级RISC机(ARM)、可编程逻辑器件以及能够执行一个或多个功能的任何电路和处理器等或其任意组合。仅为了说明，图2中的计算设备200只描述了一个处理器，但需要注意的是，本说明书中的计算设备200还可以包括多个处理器。
计算设备200的存储器(例如，只读存储器(ROM)230、随机存储器(RAM)240、硬盘270等)可以存储从车辆调配系统100的任何其他组件获取的数据/信息。示例性的ROM可以包括掩模ROM(MROM)、可编程ROM(PROM)、可擦除可编程ROM(PEROM)、电可擦除可编程ROM(EEPROM)、光盘ROM(CD-ROM)和数字通用盘ROM等。示例性的RAM可以包括动态RAM(DRAM)、双倍速率同步动态RAM(DDR SDRAM)、静态RAM(SRAM)、晶闸管RAM(T-RAM)和零电容(Z-RAM)等。
输入/输出260可以用于输入或输出信号、数据或信息。在一些实施例中，输入/输出260可以包括输入装置和输出装置。示例性输入装置可以包括键盘、鼠标、触摸屏和麦克风等或其任意组合。示例性输出装置可以包括显示设备、扬声器、打印机、投影仪等或其任意组合。示例性显示装置可以包括液晶显示器(LCD)、基于发光二极管(LED)的显示器、平板显示器、曲面显示器、电视设备、阴极射线管(CRT)等或其任意组合。
通信端口250可以连接到网络以便数据通信。连接可以是有线连接、无线连接或两者的组合。有线连接可以包括电缆、光缆或电话线等或其任意组合。无线连接可以包括蓝牙、Wi-Fi、WiMax、WLAN、ZigBee、移动网络(例如，3G、4G或5G等)等或其任意组合。在一些实施例中，通信端口250可以是标准化端口，如RS232、RS485等。在一些实施例中，通信端口250可以是专门设计的端口。
图3是根据本申请一些实施例所示的示例性移动设备300的示例性硬件和/或软件的示意图。如图3所示，移动设备300可以包括通信单元310、显示单元320、图形处理器(GPU)330、处理器(CPU)340、输入/输出单元350、内存360、存储单元370等。在一些实施例中，移动设备300也可以包括任何其它合适的组件，包括但不限于系统总线或控制器(图中未显示)。在一些实施例中，操作系统361(例如，iOS、Android、Windows Phone等)和应用程序362可以从存储单元370加载到内存360中，以便由CPU340执行。应用程序362可以包括浏览器或用于从车辆调配系统100接收文字、图像、音频或其他相关信息的应用程序。信息流的用户交互可以通过输入/输出单元350实现，并且通过网络120提供给处理设备112和/或车辆调配系统100的其他组件。
为了实现在本说明书中描述的各种模块、单元及其功能，计算设备或移动设备可以用作本说明书所描述的一个或多个组件的硬件平台。这些计算机或移动设备的硬件元件、操作系统和编程语言本质上是常规的，并且本领域技术人员熟悉这些技术后可将这些技术适应于本说明书所描述的系统。具有用户界面元 件的计算机可以用于实现个人计算机(PC)或其他类型的工作站或终端设备，如果适当地编程，计算机也可以充当服务器。
图4是根据本申请一些实施例所示的车辆调配方法400的示例性流程图。如图4所示，方法400包括下述步骤。在一些实施例中，方法400可以由处理器执行。
步骤410，获取目标租车点的关联特征数据，关联特征数据包括目标租车点和目标租车区域中的至少一种的相关特征数据；目标租车区域为与目标租车点相关的区域。
目标租车点可以是线下实地查看车辆和/或提车的租车点或车辆存放地点，例如，可以是线下租车门店、中心仓片区或特定中心仓等地点，也可以是供应或分配车辆的租赁公司。在一些实施例中，线下租车门店可以是为客户提供车辆及售后服务的场所。在一些实施例中，中心仓片区可以是包含一个或多个中心仓的划定范围区域。在一些实施例中，中心仓可以是进行车辆出入库检测、停放、调度和提车的场所，包括多个库位，库位用于存放车辆。在一些实施例中，租赁公司可以是为租车平台提供车辆的商家。
目标租车区域可以是与目标租车点相关的区域。具体地，目标租车区域可以是与目标租车点地理位置相关或业务联系相关的租车区域，例如，可以是线下租车门店、中心仓片区或特定中心仓的周边辐射划定区域，也可以互为业务关联的其中一方，例如，目标租车区域可以是线下租车门店地理位置就近或业务相关的中心仓片区或中心仓，目标租车区域可以是与中心仓地理位置就近或业务相关的中心仓片区或目标城市，目标租车区域还可以是与租赁公司业务关联(例如具有用于指定存放该租赁公司旗下车辆的关联关系等)或地理位置相关的中心仓或中心仓片区。
在一些实施例中，目标租车区域可以是与租赁公司业务关联或地理位置相关的线下租车门店，例如目标租车区域可以是与租赁公司地理位置就近的线下租车门店。
在一些实施例中，根据实际需要还可以对目标租车点与目标租车区域进行其他组合设置，即可以在线下租车门店、中心仓、中心仓片区、租赁公司、目标城市以及上述任一项的地理位置相关或业务联系相关的租车区域(该租车区域可以是根据业务需要人为划定区域，如类似周边辐射划定区域)中选取任意两者设置为目标租车点与目标租车区域。
在一些实施例中，目标租车区域可以是包括目标租车点的位置区域，例如线下租车门店的位置区域，也就是，目标租车点附近的区域，关于根据目标租车点确定该种目标租车区域的更多实施例可参见图10及相应描述，在此不再赘述。
在一些实施例中，目标租车区域可以是包括至少一个中心仓的目标城市。目标城市对应的特征数据具有该城市或者各个所选中心仓的特征，关于根据目标租车点确定该种目标租车区域的更多实施例可参见图18及相应描述，在此不再赘述。
在一些实施例中，目标租车区域可以是租赁公司对应的用于车辆存放的中心仓片区或中心仓，租赁公司一般拥有多个中心仓，根据租赁公司的租车业务情况(包括车辆丰富度、品牌影响力、爆款车型等)确定用于存放车辆的中心仓片区或中心仓需求，关于租赁公司与相应中心仓的更多实施例可参见图28及相应描述，在此不再赘述。
关联特征数据可以是目标租车点和目标租车区域中的至少一种的相关特征数据。在一些实施例中，关联特征数据可以是能够影响目标租车点或目标租车区域的租车各环节业务需求、车辆需求和/或车辆调配需求的任何可能类别的信息数据。例如，关联特征数据可以是目标租车点的被租用车辆的第一历史车辆属性信息和目标租车区域的历史租车订单对应的第二历史车辆属性信息，第一历史车辆属性信息可以包括各个被租用车辆对应的历史车型信息，第二历史车辆属性信息可以包括各历史租车订单包括的租用车辆的车型信息；关联特征数据还可以是目标租车点的环境信息和目标租车区域的选品相关特征数据，目标租车区域的选品相关特征数据可以包括目标租车区域多个维度的历史选品需求相 关数据；关联特征数据还可以是目标租车点的历史周转车辆信息及相关信息、目标租车区域的库位信息，历史周转车辆信息及相关信息可以包括历史周转车辆数目及多维度车辆周转影响数据，库位信息可以包括库位缺口数目。
第一历史车辆属性信息、第二历史车辆属性信息是指可以指示特定车辆的车型、品牌或其他车辆指示指标的车辆相应属性信息及其车辆数目。在一些实施例中，第一历史车辆属性信息可以包括目标租车点的各个被租用车辆对应的历史车型信息，第二历史车辆属性信息可以包括目标租车区域的各历史租车订单包括的租用车辆的车型信息，关于目标租车点的各个被租用车辆对应的历史车型信息、目标租车区域的各历史租车订单包括的租用车辆的车型信息的更多实施例，可参见图11及相应描述，在此不再赘述。
环境信息是指可以影响目标租车点车型需求的客观因素维度的数据。在一些实施例中，环境信息可以是反映覆盖租车区域的目标租车点的地理位置、规模等环境信息，关于环境信息的更多实施例可参见图18及相应描述，在此不再赘述。选品相关特征数据是指可以反映目标租车区域车辆选品情况的相应特征数据，选品是指确定每个目标租车点对应多个预选车型以及每个预选车型的需求量。在一些实施例中，选品相关特征数据可以包括目标租车区域多个维度的历史选品需求相关数据，目标租车区域多个维度的历史选品需求相关数据可以是目标城市对应的特征数据，关于目标城市对应的特征数据的更多实施例可参见图18及相应描述，在此不再赘述。
历史周转车辆信息及相关信息是指可以反映目标租车点历史周转车辆情况数据及可以影响周转车辆情况的其他因素数据，例如历史周转车辆信息及相关信息可以包括历史周转车辆数目及多维度车辆周转影响数据。在一些实施例中，历史周转车辆信息及相关信息可以包括租赁公司的信息维度数据、时间数据，进一步地可以包括租赁公司的总租车数量和总还车数量、历史租车单量、租赁公司规模数据、节假日信息、天气、租赁公司信用分、车型丰富度、租赁品牌知名度等维度特征，关于历史周转车辆信息及相关信息的更多实施例可参见图29、 图30及其相应描述，在此不再赘述。库位信息是指用于存放车辆的库位规划或设置情况，例如，目标租车区域的库位信息可以包括目标租车区域的库位缺口数目。在一些实施例中，目标租车区域的库位信息可以是租赁公司在每个中心仓的缺口车辆数，关于该目标租车区域的库位信息的更多实施例可参见图28至图30及其相应描述，在此不再赘述。
在一些实施例中，关联特征数据可以是目标租车点的被租用车辆的第一历史车辆属性信息和目标租车区域的选品相关特征数据，例如，关联特征数据可以是各个被租用车辆对应的历史车型信息和目标租车区域多个维度的历史选品需求相关数据。
在一些实施例中，关联特征数据可以是目标租车点的被租用车辆的第一历史车辆属性信息和目标租车区域的库位信息，例如，关联特征数据可以是各个被租用车辆对应的历史车型信息和库位缺口数目。
在一些实施例中，关联特征数据可以是目标租车点的历史周转车辆信息及相关信息和目标租车区域的历史租车订单对应的第二历史车辆属性信息，例如，关联特征数据可以是历史周转车辆数目及多维度车辆周转影响数据和各历史租车订单包括的租用车辆的车型信息。
在一些实施例中，关联特征数据可以选取以下组中的任意两种进行组合设置：目标租车点的被租用车辆的第一历史车辆属性信息、目标租车点的环境信息、目标租车点的历史周转车辆信息及相关信息、目标租车区域的历史租车订单对应的第二历史车辆属性信息、目标租车区域的选品相关特征数据、目标租车区域的库位信息。
处理器可以通过多种方式获取目标租车点的关联特征数据。例如，可以通过目标租车点上传到服务器来获取关联特征数据中的历史车辆属性信息。另外，还可以通过租车平台的服务器获取相应目标租车点的关联特征数据。
此外，还可以通过与目标租车点服务器、目标租车区域服务器具有数据通信连接关系的业务后台服务器获取相应目标租车点的关联特征数据。
步骤420，基于预测模型对目标租车点的关联特征数据进行处理，确定目标租车点的关联车辆需求，关联车辆需求为目标租车点在目标租车区域中对车辆的相关需求。
预测模型是为了通过对目标租车点的关联特征数据计算处理后获取相应关联车辆需求而采用的特定计算模型或算法，根据目标租车点的关联特征数据、期望关联车辆需求的相应特点或数据处理要求，可以采用不同的适应性计算模型或算法。例如，预测模型可以采用人工智能算法，具体地可以采用诸如决策树、随机森林、逻辑回归、支持向量机、朴素贝叶斯、K近邻算法、K均值算法、Adaboost(Boosting算法的一种)、神经网络、马尔科夫的机器学习算法。在一些实施例中，预测模型可以采用决策树算法创建得到，关于通过决策树算法建立预测模型的更多实施例可参见图19及相应描述，在此不再赘述。
在一些实施例中，步骤420可以实施为以下过程：
基于预测模型对目标租车点的被租用车辆的第一历史车辆属性信息和目标租车区域的历史租车订单对应的第二历史车辆属性信息进行处理，确定目标租车点的车型需求。其中，关联特征数据包括：目标租车点的被租用车辆的第一历史车辆属性信息和目标租车区域的历史租车订单对应的第二历史车辆属性信息；目标租车点的关联车辆需求包括目标租车点的车型需求。
在一些实施例中，步骤420可以实施为以下过程：
基于预测模型对目标租车点的环境信息和目标租车区域的选品相关特征数据进行处理，确定目标租车点的车型需求。其中，关联特征数据包括：目标租车点的环境信息和目标租车区域的选品相关特征数据；目标租车点的关联车辆需求包括目标租车点的车型需求。
在一些实施例中，步骤420可以实施为以下过程：
基于预测模型对目标租车点的历史周转车辆信息及相关信息、目标租车区域的库位信息进行处理，确定目标租车点的车辆数需求。其中，关联特征数据包括：目标租车点的历史周转车辆信息及相关信息、目标租车区域的库位信息； 目标租车点的关联车辆需求包括目标租车点的车辆数需求。
步骤430，至少基于关联车辆需求，确定目标租车点的车辆调度方案。
车辆调度方案是指可以用于指导针对目标租车点进行车辆调配操作的车辆调度相关信息数据。例如，车辆调度方案可以包括待调配的车辆信息，待调配的车辆信息可以包括待调配车辆属性信息以及每种属性车辆的数目，车辆属性信息可以包括车型、品牌或其他任何可能的车辆属性信息。
在一些实施例中，除了仅基于关联车辆需求确定目标租车点的车辆调度方案之外，还可以综合考虑车辆调配业务中统计的车辆回调情况数据、目标租车点或用户车辆特殊需求数据、目标租车区域环境变动因素数据或其他任何可能影响租车业务中车辆调配工作的相应维度数据。
在一些实施例中，上述430步骤可以实施为以下过程：
基于待调度车辆的第三车辆属性信息和目标租车点的车型需求，从待调度车辆中确定车辆调度方案的调度车辆。
第三车辆属性信息是指可以指示待调度车辆的车型、品牌或其他车辆指代指标的车辆相应属性信息。例如，第三车辆属性信息可以是待调度车辆的车型及其车辆数目信息。例如，可以基于待调度车辆的第三车辆属性信息和中心仓/线下门店的目标车型及其对应的目标车型数目，选定与中心仓/线下门店的目标车型匹配、目标车型数目的调度车辆。
在一些实施例中，上述430步骤可以实施为以下过程：
基于待调度车辆的第三车辆属性信息和目标租车点的车辆数需求，从待调度车辆中确定车辆调度方案的调度车辆。例如，可以基于待调度车辆的第三车辆属性信息和租赁公司的车辆数配额需求，从待调度车辆中确定车辆调度方案的调度车辆。
本申请实施例提供的车辆调配方法、装置、设备及计算机可读存储介质，通过获取目标租车点影响车辆调配业务的关联特征数据，利用预测模型处理确定目标租车点的关联车辆需求，然后再根据该关联车辆需求确定其车辆调度方 案，提供了一种调配更合理、满足多种租车业务场景需求的高效车辆调配方案节约业务运营成本，满足了用户多重需求，提高了用户体验。
图5是根据本申请一些实施例所示的车辆调配装置500的示例性结构图。
基于同一发明构思，本申请实施例一些实施例提供了一种车辆调配装置500，车辆调配装置500包括获取模块501、第一确定模块502和第二确定模块503。
获取模块501，用于：获取目标租车点的关联特征数据，关联特征数据包括目标租车点和目标租车区域中的至少一种的相关特征数据；目标租车区域为与目标租车点相关的区域。关于获取目标租车点的关联特征数据的更多细节参见图4及相关描述，在此不再赘述。
在一些实施例中，获取模块501可以用于获取目标租车点的位置信息以及在所述目标租车点被租用车辆的历史车辆属性信息并确定与所述目标租车点对应的位置区域，以及获取定位信息属于所述位置区域的用户端对应的历史租车订单。
在一些实施例中，获取模块501可以用于获取目标城市在第一预设历史时间段内的特征数据，所述目标城市为设有至少一个所述中心仓的城市，所述特征数据用于表示所述目标城市中各个中心仓的环境信息以及所在目标城市的车型的相关联数据。
在一些实施例中，获取模块501可以用于负责获取各方来源数据，并做数据清洗、特征处理和结构化存储。
第一确定模块502，用于：基于预测模型对目标租车点的关联特征数据进行处理，确定目标租车点的关联车辆需求，关联车辆需求为目标租车点在目标租车区域中对车辆的相关需求。关于基于预测模型确定目标租车点的关联车辆需求的更多细节参见图4及相关描述，在此不再赘述。
在一些实施例中，第一确定模块502可以用于：通过预测模型进行租车需求预测和还车数量预测，为后续模块提供决策数据支撑。
第二确定模块503，用于：至少基于关联车辆需求，确定目标租车点的车辆调度方案。关于确定目标租车点的车辆调度方案的更多细节参见图4及相关描述，在此不再赘述。可以用于其基于预测的数据以及业务场景，做自动化的实时决策。
在一些实施例中，第二确定模块503可以用于基于预测的数据以及业务场景，做自动化的实时决策。
需要注意的是，以上对于车辆调配装置500及其模块的描述，仅为描述方便，并不能把本申请限制在所举实施例范围之内。可以理解，对于本领域的技术人员来说，在了解该系统的原理后，可能在不背离这一原理的情况下，对各个模块进行任意组合，或者构成子系统与其他模块连接。在一些实施例中，图5中披露的获取模块501、第一确定模块502、第二确定模块503可以是一个系统中的不同模块，也可以是一个模块实现上述的两个或两个以上模块的功能。例如，各个模块可以共用一个存储模块，各个模块也可以分别具有各自的存储模块。诸如此类的变形，均在本申请的保护范围之内。
图6是根据本申请一些实施例所示的车辆调配设备600的示例性结构图。
基于同一发明构思，本申请一些实施例还提供了一种车辆调配设备600，车辆调配设备600包括存储器601、处理器602和计算机程序，计算机程序存储在存储器601中，并被配置为由处理器602执行上述任一实施例提供的车辆调配方法，关于执行的车辆调配方法流程具体细节参见上述方法实施例，在此不再赘述。
在上述的图6所示的实施例中，应理解，处理器可以是中央处理单元(英文：Central Processing Unit，简称：CPU)，还可以是其他通用处理器、数字信号处理器(英文：Digital Signal Processor，简称：DSP)、专用集成电路(英文：Application Specific Integrated Circuit，简称：ASIC)等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。结合发明所公开的方法的步骤可以直接体现为硬件处理器执行完成，或者用处理器中的硬件及软件模块组 合执行完成。
存储器可能包含高速RAM存储器，也可能还包括非易失性存储NVM，例如至少一个磁盘存储器。
总线可以是工业标准体系结构(Industry Standard Architecture，ISA)总线、外部设备互连(Peripheral Component Interconnect，PCI)总线或扩展工业标准体系结构(Extended Industry Standard Architecture，EISA)总线等。总线可以分为地址总线、数据总线、控制总线等。为便于表示，本申请附图中的总线并不限定仅有一根总线或一种类型的总线。
另外，本申请实施例还提供一种计算机可读存储介质，其上存储有计算机程序，所述计算机程序被处理器执行以实现上述实施例所述的车辆调配方法。
上述的计算机可读存储介质，上述可读存储介质可以是由任何类型的易失性或非易失性存储设备或者它们的组合实现，如静态随机存取存储器(SRAM)，电可擦除可编程只读存储器(EEPROM)，可擦除可编程只读存储器(EPROM)，可编程只读存储器(PROM)，只读存储器(ROM)，磁存储器，快闪存储器，磁盘或光盘。可读存储介质可以是通用或专用计算机能够存取的任何可用介质。
一种示例性的可读存储介质耦合至处理器，从而使处理器能够从该可读存储介质读取信息，且可向该可读存储介质写入信息。当然，可读存储介质也可以是处理器的组成部分。处理器和可读存储介质可以位于专用集成电路(Application Specific Integrated Circuits，简称：ASIC)中。当然，处理器和可读存储介质也可以作为分立组件存在于设备中。
图7是根据本申请一些实施例所示的车辆调配方法700的示例性流程图。在一些实施例中，方法700可以由处理器执行。
步骤710，获取目标租车点的关联特征数据，关联特征数据包括：目标租车点的被租用车辆的第一历史车辆属性信息和目标租车区域的历史租车订单对应的第二历史车辆属性信息。
步骤720，基于预测模型对目标租车点的被租用车辆的第一历史车辆属性 信息和目标租车区域的历史租车订单对应的第二历史车辆属性信息进行处理，确定目标租车点的车型需求。
步骤730，至少基于目标租车点的车型需求，确定目标租车点的车辆调度方案。
本实施例提供的车辆调度方法，通过获取目标租车点的关联特征数据，包括：被租用车辆的第一历史车辆属性信息和目标租车区域的历史租车订单对应的第二历史车辆属性信息，基于预测模型对关联特征数据进行处理，确认目标租车点的车型需求，进一步确认目标租车点的调度方案，可以精确化目标租车点的车型需求，提高车辆调度与实际车型需求的匹配度，进而提高了车辆调配效率，有利于租车点租车效率的提升。
图8是根据本申请一些实施例所示的车辆调配方法800的示例性流程图。在一些实施例中，方法800可以由处理器执行。
步骤810，获取目标租车点的关联特征数据，关联特征数据包括：目标租车点的被租用车辆的第一历史车辆属性信息和目标租车区域的历史租车订单对应的第二历史车辆属性信息；
步骤820，基于预测模型对目标租车点的被租用车辆的第一历史车辆属性信息和目标租车区域的历史租车订单对应的第二历史车辆属性信息进行处理，确定目标租车点的车型需求；
步骤830，基于待调度车辆的第三车辆属性信息和目标租车点的车型需求，从待调度车辆中确定车辆调度方案的调度车辆。
图9是根据本申请一些实施例所示的车辆调配方法900的示例性流程图。在一些实施例中，方法900可以由处理器执行。
步骤910，获取目标租车点的关联特征数据，关联特征数据包括：目标租车点的被租用车辆的各个被租用车辆对应的历史车型信息和目标租车区域的历史租车订单对应的历史租车订单包括的租用车辆的车型信息；
步骤920，基于预测模型对目标租车点各个被租用车辆对应的历史车型信 息、各历史租车订单包括的租用车辆的车型信息进行处理，确定目标租车点的目标车型及其对应的目标车型数目；
步骤930，基于待调度车辆的第三车辆属性信息和目标租车点的目标车型及其对应的目标车型数目，选定与目标租车点的目标车型匹配、目标车型数目的调度车辆。
本实施例提供的车辆调度方法，通过获取包括各个被租用车辆对应的历史车型信息、各历史租车订单包括的租用车辆的车型信息在内的关联特征数据，基于预测模型对关联特征数据进行处理，确定目标租车点的目标车型即其对应的数目，并基于此为目标租车点确定调度车辆。该方法使用户可以直接在租车点看车、提车而无需去目标租车区域，满足了用户就近看车提车的需求，节约了用户时间，提升了用户的租车体验。用户带到租车点看车、提车也提升了租车点的访问率，提高了租车点的租车效率。
本申请一些实施例提供了一种车辆调配方法及其硬件装置。在一些实施例中，目标租车点的关联特征数据可以选取目标租车点被租用车辆的历史车辆属性信息和目标租车点对应的位置区域(也即目标租车区域)的历史租车订单对应的租用车辆属性信息，待调度车辆的第三车辆属性信息可以选取预设的待调度车辆的车辆属性信息，通过预测模型处理，确定目标租车点的车辆调度方案。
在一些实施例中，在获取目标租车点的位置信息和该目标租车点被租用车辆的历史车辆属性信息后，确定与目标租车点对应的位置区域，获取定位信息属于该位置区域的用户的历史租车订单，通过预测模型对历史车辆属性信息和历史租车订单对应的租用车辆属性信息进行处理，获取目标租车点的关联车辆需求；然后根据待调度车辆的车辆属性信息，从待调度车辆中，为目标租车点确定调度车辆，并将确定的车辆调度到目标租车点，这样，提高了目标租车点的线下访问率，也提高了目标租车点的租车效率，同时，用户能够在目标租车点直接提车，而无需去中心仓看车提车，节约了用户时间，提高了用户的体验，提高了平台的服务质量。
请参见图10，示出了本申请一些实施例提供的一种车辆调配方法1000的示例性流程图，方法1000可以由处理器来实现。
步骤1001，获取目标租车点的位置信息，以及在所述目标租车点被租用车辆的历史车辆属性信息；
步骤1002，基于所述目标租车点的位置信息，确定与所述目标租车点对应的位置区域；
步骤1003，获取定位信息属于所述位置区域的用户端对应的历史租车订单；
步骤1004，通过预测模型对所述历史车辆属性信息和所述历史租车订单对应的租用车辆属性信息进行处理，根据预设的待调度车辆的车辆属性信息，从多个待调度车辆中，为所述目标租车点确定调度车辆，并将确定的调度车辆的信息发送至调度终端，从而能够将确定的车辆调度到所述目标租车点。
在步骤1001中，目标租车点为用户可以线下实地查看车辆并提车的租车点；位置信息可以通过全球定位系统(Global Positioning System，GPS)坐标进行表示；目标租车点被租用过的车辆可以为在历史时间段，用户在目标租车点租用的车辆，比如，线下下单的方式在目标租车点租用过的车辆，或者通过目标租车点的租车系统下单方式租用的车辆；历史车辆属性信息包括车辆的车牌号、车型信息、车辆行驶信息、车辆维护信息等，其中，车型信息可以为多用途汽车(Multi-Purpose Vehicle，MPV)车型、运动型实用汽车(Sport Utility Vehicle，SUV)、大型车、中等车、小型车等；车辆行驶信息包括车辆行驶里程，比如，车辆累计行驶100公里；车辆维护信息包括车辆的年检信息、维修信息、保险信息等信息；历史车辆属性信息可以为目标租车点上传到服务器的。
在步骤1002中，位置区域为包括目标租车点的区域，也就是，目标租店点附近的区域，该位置区域可以为矩形区域，可以为圆形区域，也可以是预先划分好的区域。在位置区域时预先划分好的区域时，可以按照行政功能对城市进行划分得到，也可以基于租车点的周边信息，得到位置区域。
在确定位置区域时，可以通过以下任意一种方式得到目标租车点对应的位置区域：
方式一：在基于所述目标租车点的位置信息，确定与所述目标租车点对应的位置区域时，可以包括以下步骤：
以所述目标租车点的位置信息为圆心、以距离阈值为半径形成的圆形区域作为与所述目标租车点对应的位置区域。
这里，预设距离阈值可以为预先设置的，例如，距离阈值可以为5公里，此处需要注意，预设距离阈值可以根据实际情况确定。
在具体实施过程中，将目标租车点的位置信息作为圆心，将距离阈值作为半径，形成一个圆形区域，将该圆形区域作为目标租车点对应的位置区域。
方式二：在基于所述目标租车点的位置信息，确定与所述目标租车点对应的位置区域时，可以包括以下步骤：
根据所述目标租车点的位置信息，以及预设的位置信息范围和位置区域之间的对应关系，确定与所述目标租车点对应的位置区域。
这里，预设的位置区域可以为根据区域的行政功能划分的。
在具体实施过程中，在获取目标租车点的位置信息后，比对位置信息和预设的多个位置信息范围，若存在位置信息范围包括目标租车点的位置信息，则将存在的位置信息范围对应的位置区域确定为目标租车点对应的位置区域。
在步骤1003中，定位信息为用户在线上浏览租车平台的车辆时使用的定位位置，比如用户使用用户端在线上浏览车辆时用户端的定位信息，该定位信息可以通过GPS坐标表示；历史租车订单中包括用户租用车辆的车辆属性信息，比如，租用车辆的车型信息、车辆行驶信息、车辆维护信息等，具体内容可参考上文。
在步骤1004中，待调度车辆为租车平台对车辆供应商的车辆进行筛选得到的，待调度车辆中包括不同车型分别对应的车辆，租车平台在筛选车辆时，通常是通过车辆供应商上报车辆数目、车辆属性信息以及车辆报价，租车平台根据 车辆报价、车型等信息，确定需要的多种车型，以及每种车型对应的车辆数目；待调度车辆的车辆属性信息包括待调度车辆的车型，以及对应的总数目。
请参见图11，示出了本申请一些实施例提供的一种车辆调配方法1100的示例性流程图，方法1100可以由处理器来实现。
步骤1101，通过预测模型处理所述目标租车点各个被租用车辆对应的历史车型信息，确定所述目标租车点对应的第一车型集合，以及所述第一车型集合中不同车型分别对应的被租用车辆的第一数目；
步骤1102，通过预测模型处理各所述历史租车订单包括的租用车辆的车型信息，确定所述位置区域对应的第二车型集合，以及第二车型集合中不同车型分别对应的租用车辆的第二数目；
步骤1103，通过预测模型处理所述第一车型集合中不同车型分别对应的第一数目，以及所述第二车型集合中不同车型分别对应的第二数目，从所述第一车型集合和所述第二车型集合中，确定目标车型，并确定所述目标车型对应的第三数目；
步骤1104，基于所述待调度车辆的车辆属性信息，从所述多个待调度车辆中，选择与所述目标车型匹配的、第三数目的待调度车辆。
在步骤1101中，第一车型集合中包括多个车型，该第一车型集合表征目标租车点被租用车辆的车型，例如，目标租车点被租用过的车辆为100辆，其中，A车型20辆，B车型70辆，C车型10辆，则第一车型集合中的车型为A、B、C；第一数目为第一车辆集合中的车型对应的车辆数目，比如，车型A对应20辆，则车型A对应的第一数目为20；第二车型集合为目标租车点对应的位置区域中被租用过的车辆的车型，第二数目为第二车辆集合中的车型对应的车辆数目，可参考上述第一车型集合中的示例。
在具体实施过程中，在获取目标租车点被租用过的各个车辆的车型信息后，统计目标租车点存在的车型，生成第一车型集合，统计第一车型集合中的每个车型对应的车辆数目，也就是，每个车型对应的被租用车辆的第一数目。
例如，目标租车点存在5辆被租用车辆，分别为C1、C2、C3、C4和C5，C1对应的车型为A1，C2对应的车型为A2，C3对应的车型为A1，C4对应的车型为A3，C5对应的车型为A2，则第一车型集合中包括车型A1、A2和A3，其中车型A1对应的第一数目为2，车型A2对应的第一数目为2，车型A3对应的第一数目为1。
在获取目标租车点所处的位置区域的用户的历史租车订单后，根据每个历史租车订单中包括的租用车辆的车型信息(如车型)，统计目标租车点所对应的位置区域存在的车型，生成第二车型集合，统计第二车型集合中的每个车型对应的车辆数目，也就是，每个车型对应被租用车辆的第二数目。
例如，目标租车点所在的位置区域存在5个历史租车订单，历史租车订单中包括的租用车辆分别为C6、C7、C8、C9和C10，C6对应的车型为A1，C7对应的车型为A5，C8对应的车型为A1，C9对应的车型为A3，C10对应的车型为A1，则第二车型集合中包括车型A1、A3和A5，其中车型A1对应的第一数目为3，车型A3对应的第一数目为1，车型A5对应的第一数目为1。
在确定第一车型集合中每种车型对应的第一数目，和第二车型集合中的每个车型对应的第二数目后，确定第一车型集合和第二车型集合中的相同车型，从相同车型对应的第一数目和第二数目中选择最大数目，按照数目由大到小的顺序对第一车型集合和第二车型集合中全部车型进行排序，将排序靠前的前预设数目个车型确定为目标车型，并确定每个目标车型对应的第三数目。其中，预设数目可以根据实际情况确定。
若目标车型为第一车型集合和第二车型集合中共同包含的车型，则第三数目为相同车型对应的第一数目和第二数目中的最大数目，若目标车型仅为第一车型集合中的车型，则第三数目为该车型在第一车型集合中对应的第一数目，若目标车型仅为第二车型集合中的车型，则第三数目为该车型在第二车型集合中对应的第二数目。
例如，第一车型集合中包括车型A1、A2和A3，其中车型A1对应的第 一数目为2，车型A2对应的第一数目为2，车型A3对应的第一数目为1，第二车型集合中包括车型A1、A3和A5，其中车型A1对应的第一数目为2，车型A3对应的第一数目为2，车型A5对应的第一数目为1，第一车型集合和第二车型集合共同包含的车型为A1和A3，则车型A1对应的被租用车辆的数目为2，A3对应的被租用车辆的数目为2，按照数目由大到小的顺序排序后的车型顺序为A1和A3、A2、A5，目标车型对应的预设数目为3，则目标车型为A1、A3、A2，A1对应的第三数目为2，A3对应的第三数目为3，A2对应的第三数目为2。
针对确定的每个目标车型，比对该目标车型与待调度车辆的车型信息，若存在待调度车辆的车型信息与目标车型匹配，也就是，存在待调度车辆的车型信息与目标车型一致，则获取存在的待调度车辆的车辆总数。
若目标车型对应的第三数目小于对应待调度车辆的车辆总数，则选择第三数目的待调度车辆；若目标车型对应的第三数目大于对应待调度车辆的车辆总数，则可以为目标租车点调度车辆总数的待调度车辆，并计算第三数目与车辆总数之间的差值数目，等待中心仓具有该差值数目的待调度车辆后，再为该目标租车点分配差值数目的待调度车辆。
例如，待调度车辆的车型信息包括A1、A2、A3、A4、A5，其中，车型A1对应的车辆总数为10，车型A2对应的车辆总数为10，车型A3对应的车辆总数为10，车型A4对应的车辆总数为10，车型A5对应的车辆总数为10，目标车型包括A1、A3、A2，A1对应的第三数目为4，A3对应的第三数目为3，A2对应的第三数目为2，则为目标租车点分配4辆A1车型的车辆，为目标租车点分配3辆A3车型的车辆，为目标租车点分配2辆A2车型的车辆。
在向目标租车点调度车辆时，目标租车点需要的调度车辆可能与另一个租车点存在冲突，例如，目标租车点需要调度的A型车辆的数目为M，另一个租车点需要调度的A型车辆为N，而A型待调度车辆的总数小于M+N，如何在保证租车评价资源最大化利用的前提下为存在调度冲突的租车点调度车辆成为 需要解决的一个问题。
在为冲突租车点调度车辆时，也就是，基于所述待调度车辆的车辆属性信息，从所述多个待调度车辆中，选择与所述目标车型匹配的、第三数目的待调度车辆，可以包括以下步骤：
若确定存在另一个租车点对应的调度车辆的车型信息与所述目标车型匹配，且车型信息与所述目标车型匹配的调度车辆的车辆数目与所述第三数目的和值，大于目标车型对应的待调度车辆的总数，则比对所述目标租车点和另一个租车点的优先级；
若目标租车点的优先级高于另一个租车点的优先级，则将与所述目标车型匹配的、第三数目的待调度车辆确定为所述目标租车点确定的调度车辆。
这里，优先级表征租车点的重要程度，该重要程度越高表征租车点的优先级越高，优先级可以是根据租车点的资源价值确定的，也可以是根据租车点的服务价值确定的，可以根据实际情况确定。
在具体实施过程中，在为各个租车点调度车辆时，向某一个租车点调度的车辆的车型信息中，可能存在至少一种车型与对应的目标车型匹配，且至少一种车型对应的调度车辆的数量数目与对应目标车型所对应的第三数目的和值大于对应目标车型所对应的待调度车辆的总数，也就是，针对另一个租车点对应的每种车型，计算该种车型对应的车辆数目与对应的目标车型的第三数目的和值，若该和值大于目标车辆对应的待调度车辆总数目，则比对目标租车点和另一个租车点的优先级。
若目标租车点的优先级高于另一个租车点的优先级，则将与目标车型匹配的、第三数目的待调度车辆调度到目标租车点；若目标租车点的优先级低于另一个租车点的优先级，则优先为另一个租车点分配车辆，若目标租车点的优先级等于另一个租车点的优先级，则可以平均为两个租车点分配车辆，即，计算目标车型对应的待调度车辆的总数与租车点数目的比值，为每个租车点分配与该比值相同数目的待调度车辆。
例如，为目标租车点S1分配4辆A1型的车辆，为另一租车点S2分配了5辆A1车型的车辆，目标租车点S1的优先级为L1，另一租车点S2的优先级为L2，其中，目标租车点S1和另一租车点S2存在的共同的车型为A1型，目标租车点S1和另一租车点S2对A1型车辆的总需求量为10辆，而A型的待调度车辆的车辆总数为8辆，目标租车点S1和另一租车点S2对A1型车辆的总需求量大于A1型待调度车辆的总数，此时，比对目标租车点S1的优先级L1和另一租车点S2的优先级L2，若目标租车点S1的优先级L1大于另一租车点S2的优先级L2，则优先为目标租车点S1分配4辆A1型车辆，若目标租车点S1的优先级L1小于另一租车点S2的优先级L2，则可以优先为另一租车点S2分配6辆A1型车辆，若目标租车点S1的优先级L1等于另一租车点S2的优先级L2，则可以为目标租车点分配4辆A1型车辆，为另一个租车点分配4辆A1型车辆。
除了考虑目标租车点被租用车辆的车辆属性信息外，还可以考虑目标租车点上报的车辆需求信息或访问目标租车点的用户兴趣信息，为目标租车点确定需要调度的车辆。以下分别介绍。
在从多个待调度车辆中，为目标租车点确定调度车辆时，考虑目标租车点的车辆需求信息和历史车辆属性信息为目标租车点确定调度车辆，可以包括以下步骤：
获取所述目标租车点对应的车辆需求信息；
基于所述历史车辆属性信息、所述历史租车订单对应的租用车辆属性信息和所述车辆需求信息，以及预设的待调度车辆的车辆属性信息，从多个待调度车辆中，为所述目标租车点确定调度车辆。
这里，车辆需求信息包括车型信息，以及每个车型对应的需求数目。
在具体实施过程中，通过目标租车点被租用过的车辆的历史车辆属性信息(车型信息)，确定目标租车点对应的第一车型集合，以及第一车型集合中的每种车型对应的第一数目，通过目标租车点所处的位置区域的用户的历史租车 订单，确定目标租车点所处的位置区域对应的第二车型集合，以及第二车型集合中的每种车型对应的第二数目。其中，第一车型集合、第一数目、第二车型集合和第二数目的确定过程，可以参考上文，此处不进行赘述。
获取目标租车点的需求车辆对应的车型，以及对应的车辆数目，若第一车型集合和第二车型集合中存在相同的车型，且相同的车型与目标租车点需求车辆的车型匹配，则比对第一车型集合中该车型的数目、第二车型集合中该车型的数目以及需求车辆对应的该车型的数量数目，选择最大数目，并给目标租车点分配最大数目的该车型的待调度车辆。
若仅第一车型集合中存在车型与目标租车点需求车辆的任一车型匹配，则从任一车型对应的车辆数目和第一车型集合中存在车型对应的第一数目中选择最大数目，并给目标租车点分配该最大数目对应的车型的待调度车辆。
若仅第二车型集合中存在车型与目标租车点需求车辆的任一车型匹配，则从任一车型对应的车辆数目和第二车型集合中存在车型对应的第二数目中选择最大数目，并给目标租车点分配该最大数目对应的任一车型的待调度车辆。
若第一车型集合和第二车型集合中，均不存在与目标租车点需求车辆的车型匹配的车型，且第一数目、第二数目和需求车辆的车辆数目均小于对应车型的待调度车辆的总数时，那么，从第一车型集合、第二车型集合中、以及需求车辆对应的车型中，随机选择车型作为目标车型，并将随机选择的车型对应的数目作为目标车型对应的待调度车辆的数目，比如，选择第一车型集合中的车型A作为目标车型，则将车型A对应的第一数目确定为需要调度的A型车辆的车辆数目。
考虑访问目标租车点的用户兴趣信息和历史车辆属性信息，在从多个待调度车辆中，为目标租车点确定调度车辆时，可以包括以下步骤：
获取访问所述目标租车点的用户兴趣信息；
基于所述历史车辆属性信息、所述历史租车订单对应的租用车辆属性信息和所述用户兴趣信息，以及预设的待调度车辆的车辆属性信息，从多个待调度 车辆中，为所述目标租车点确定调度车辆。
这里，用户兴趣信息表征用户感兴趣的车型信息，该用户兴趣信息可以为基于用户的历史浏览数据确定的，也可以是基于用户标注的兴趣确定的，可以根据实际情况确定的。
在具体实施过程中，第一车型集合以及第一车型集合中的每种车型对应的第一数目、第二车型集合以及第二车型集合中的每种车型对应的第二数目的确定过程，可以参考上文，此处不进行赘述。
获取访问目标租车点的用户兴趣信息，确定用户的感兴趣车型，比对感兴趣车型、第一车型集合中车型以及第二车型集合中的车型，若第一车型集合中和第二车型集合中存在相同的车型，且相同的车型与感兴趣车型匹配，那么，确定与感兴趣车型对应的第一数目和第二数目，选择最大数目，并给目标租车点分配最大数目的感兴趣车型的待调度车辆。
若仅第一车型集合中存在车型与感兴趣车型匹配，则给目标租车点分配第一数目的感兴趣车型的待调度车辆。
若仅第二车型集合中存在车型与感兴趣车型匹配，则给目标租车点分配第二数目的感兴趣车型的待调度车辆。
若第一车型集合和第二车型集合中，均不存在与感兴趣车型匹配的车型，那么，从第一车型集合、第二车型集合中，随机选择车型作为目标车型，并将随机选择的车型对应的数目作为目标车型对应的待调度车辆的数目，比如，选择第一车型集合中的车型A作为目标车型，则将车型A对应的第一数目确定为需要调度的A型车辆的车辆数目，并记录目标兴趣点的感兴趣车型，以便下次分配时考虑。
由于车辆在向目标租车点调度过程(也就是，将车辆由中心仓运输到目标租车点)，可能存在运输问题导致调度到目标租车点的车辆存在异常问题，如，车体存在划痕等，或者目标租车点的车辆出现滞租问题，也就是，目标租车点的车辆在一定时间段内并没有被用户租走，为了提高目标租车点的租车效率，可以 对目标租车点的车辆进行回调。
在对目标租车点的车辆进行回调时，可以包括以下步骤：
获取所述目标租车点的回调车辆订单；所述回调车辆订单中包括回调车辆的历史服务信息；
若基于所述回调车辆的历史服务信息，确定所述回调车辆满足回调条件，则指示所述目标租车点控制所述回调车辆进行回调。
这里，历史服务信息包括回调车辆的车牌号、车型信息、车辆数目、车体信息、年检信息、使用年限、行驶里程等信息，其中，车辆数目为每种车型对应的车辆的数目，车体信息用于表征车体是否存在划伤，年检信息用于表征车辆是否进行年检，使用年限表征车辆使用的时长，行驶里程表征车型行驶的总公里数。
在具体实施过程中，在接收到目标租车点的回调车辆订单后，基于回调车辆的车体信息、年检信息、使用年限和行驶里程等信息，确定车辆是否满足回调条件。
回调条件包括以下条件中的至少一个条件：
车体存在划痕；距离进行年检的时间小于预设时长；使用年限大于预设年限阈值；行驶里程大于预设里程阈值。其中，预设时长可以为一个月，预设年限阈值可以根据车辆报废年限确定，里程阈值可以根据历史车辆行驶里程确定。
在确定车辆满足回调条件中的至少一个条件后，则指示目标租车点将回调车辆订单中的车辆回调到车辆回收中心。
在指示所述目标租车点控制所述回调车辆进行回调时，为了减少运输车辆的费用，从多个车辆回收中心中，为目标租车点选择距离最近的车辆回收中心，可以包括以下步骤：
基于所述目标租车点的位置信息，以及预设的多个车辆回收中心的位置信息，确定所述目标租车点分别与每个所述车辆回收中心之间的行驶距离；
将最短行驶距离对应的车辆回收中心确定为接收所述回调车辆的车辆回收中心；
指示所述目标租车点将所述回调车辆调度到确定的车辆回收中心。
这里，车辆回收中心的位置信息可以通过GPS坐标进行表示。
在具体实施过程中，在确定对目标租车点的车辆进行回调后，针对预设的每个车辆回收中心，根据目标租车点的位置信息，和该车辆回收中心的位置信息，计算目标租车点到车辆回收中心的行驶距离。其中，可以利用地图信息，计算目标租车点到车辆回收中心之间的行驶距离，本申请对此不予限制。
在确定目标租车点到各个车辆回收中心的行驶距离后，选择最短行驶距离对应的车辆回收中心作为接收回调车辆的车辆回收中心，指示目标租车点将回调车辆运输到确定的车辆回收中心，已完成车辆回收，便于对回收车辆的重新利用，提高回收车辆的利用率。
例如，目标租车点为S1，车辆回收中心为B1、B2、B3，S1到B1的行驶距离为L1，S1到B2的行驶距离为L2，S1到B3的行驶距离为L3，其中L3为最小行驶距离，则确定B3为接收回调车辆的车辆回收中心，将目标租车点的回调车辆运输到B3。
基于同一发明构思，本申请实施例中还提供了车辆调配方法对应的车辆调配装置，由于本申请实施例中的方法解决问题的原理与本申请实施例上述车辆调配方法相似，因此装置的实施可以参见方法的实施，重复之处不再赘述。
本申请实施例提供了一种车辆调配装置，如图12所示，该装置1200包括：
第一获取模块1201，用于获取目标租车点的位置信息，以及在所述目标租车点被租用车辆的历史车辆属性信息；
确定模块1202，用于基于所述目标租车点的位置信息，确定与所述目标租车点对应的位置区域；
第二获取模块1203，用于获取定位信息属于所述位置区域的用户端对应的历史租车订单；
处理模块1204，用于通过预测模型对所述历史车辆属性信息和所述历史 租车订单对应的租用车辆属性信息进行处理，根据预设的待调度车辆的车辆属性信息，从多个待调度车辆中，为所述目标租车点确定调度车辆，并将确定的调度车辆的信息发送至调度终端，从而能够将确定的车辆调度到所述目标租车点。
如前所述获取模块501可以获取目标租车点的位置信息以及在所述目标租车点被租用车辆的历史车辆属性信息并确定与所述目标租车点对应的位置区域，以及获取定位信息属于所述位置区域的用户端对应的历史租车订单。在一些实施例中获取模块501可以包括多个子模块，例如，获取模块501可以包括第一获取模块1201、确定模块1202和第二获取模块1203。在一些实施例中，这里的处理模块1204可以包括前述的第一确定模块502和第二确定模块503。
在一些实施例中，所述处理模块1204根据以下步骤从多个待调度车辆中，为所述目标租车点确定需要调度的车辆：
通过预测模型处理所述目标租车点各个被租用车辆对应的历史车型信息，确定所述目标租车点对应的第一车型集合，以及所述第一车型集合中不同车型分别对应的被租用车辆的第一数目；
通过预测模型处理各所述历史租车订单包括的租用车辆的车型信息，确定所述位置区域对应的第二车型集合，以及第二车型集合中不同车型分别对应的租用车辆的第二数目；
通过预测模型处理所述第一车型集合中不同车型分别对应的第一数目，以及所述第二车型集合中不同车型分别对应的第二数目，从所述第一车型集合和所述第二车型集合中，确定目标车型，并确定所述目标车型对应的第三数目；
基于所述待调度车辆的车辆属性信息，从所述多个待调度车辆中，选择与所述目标车型匹配的、第三数目的待调度车辆。
在一些实施例中，所述处理模块1204还用于：
若确定存在另一个租车点对应的调度车辆的车型信息与所述目标车型匹配，且车型信息与所述目标车型匹配的调度车辆的车辆数目与所述第三数目的和值，大于目标车型对应的待调度车辆的总数，则比对所述目标租车点和另一个 租车点的优先级；
若目标租车点的优先级高于另一个租车点的优先级，则将与所述目标车型匹配的、第三数目的待调度车辆确定为所述目标租车点确定的调度车辆。
在一些实施例中，所述处理模块1204具体用于：
获取所述目标租车点对应的车辆需求信息；
基于所述历史车辆属性信息、所述历史租车订单对应的租用车辆属性信息和所述车辆需求信息，以及预设的待调度车辆的车辆属性信息，从多个待调度车辆中，为所述目标租车点确定调度车辆。
在一些实施例中，所述处理模块1204具体用于：
获取访问所述目标租车点的用户兴趣信息；
基于所述历史车辆属性信息、所述历史租车订单对应的租用车辆属性信息和所述用户兴趣信息，以及预设的待调度车辆的车辆属性信息，从多个待调度车辆中，为所述目标租车点确定调度车辆。
在一些实施例中，所述第一获取模块1201还用于：
获取所述目标租车点的回调车辆订单；所述回调车辆订单中包括回调车辆的历史服务信息；
所述确定模块1202还用于：
若基于所述回调车辆的历史服务信息，确定所述回调车辆满足回调条件，则指示所述目标租车点控制所述回调车辆进行回调。
在一些实施例中，所述确定模块1202还用于根据以下步骤指示所述目标租车点控制所述回调车辆进行回调：
基于所述目标租车点的位置信息，以及预设的多个车辆回收中心的位置信息，确定所述目标租车点分别与每个所述车辆回收中心之间的行驶距离；
将最短行驶距离对应的车辆回收中心确定为接收所述回调车辆的车辆回收中心；
指示所述目标租车点将所述回调车辆调度到确定的车辆回收中心。
在一些实施例中，所述确定模块1202具体用于：
以所述目标租车点的位置信息为圆心、以距离阈值为半径形成的圆形区域作为与所述目标租车点对应的位置区域。
本申请实施例还提供了一种电子设备1300，如图13所示，为本申请实施例提供的电子设备1300结构示意图，包括：处理器1301、存储器1302、和总线1303。所述存储器1302存储有所述处理器1301可执行的机器可读指令(比如，图12中的装置中第一获取模块1201、确定模块1202、第二获取模块1203和处理模块1204对应的执行指令等)，当电子设备1300运行时，所述处理器1301与所述存储器1302之间通过总线1303通信，所述机器可读指令被所述处理器1301执行时执行如下处理：
获取目标租车点的位置信息，以及在所述目标租车点被租用车辆的历史车辆属性信息；
基于所述目标租车点的位置信息，确定与所述目标租车点对应的位置区域；
获取定位信息属于所述位置区域的用户端对应的历史租车订单；
基于所述历史车辆属性信息和所述历史租车订单对应的租用车辆属性信息，以及预设的待调度车辆的车辆属性信息，从多个待调度车辆中，为所述目标租车点确定调度车辆，并将确定的车辆调度到所述目标租车点。
一种可能的实施方式中，处理器1301执行的指令中，基于所述历史车辆属性信息和所述历史租车订单对应的租用车辆属性信息，以及预设的待调度车辆的车辆属性信息，从多个待调度车辆中，为所述目标租车点确定调度车辆，包括：
基于所述目标租车点各个被租用车辆对应的历史车型信息，确定所述目标租车点对应的第一车型集合，以及所述第一车型集合中不同车型分别对应的被租用车辆的第一数目；
基于各所述历史租车订单包括的租用车辆的车型信息，确定所述位置区 域对应的第二车型集合，以及第二车型集合中不同车型分别对应的租用车辆的第二数目；
基于所述第一车型集合中不同车型分别对应的第一数目，以及所述第二车型集合中不同车型分别对应的第二数目，从所述第一车型集合和所述第二车型集合中，确定目标车型，并确定所述目标车型对应的第三数目；
基于所述待调度车辆的车辆属性信息，从所述多个待调度车辆中，选择与所述目标车型匹配的、第三数目的待调度车辆。
一种可能的实施方式中，处理器1301执行的指令中，基于所述待调度车辆的车辆属性信息，从所述多个待调度车辆中，选择与所述目标车型匹配的、第三数目的待调度车辆，包括：
若确定存在另一个租车点对应的调度车辆的车型信息与所述目标车型匹配，且车型信息与所述目标车型匹配的调度车辆的车辆数目与所述第三数目的和值，大于目标车型对应的待调度车辆的总数，则比对所述目标租车点和另一个租车点的优先级；
若目标租车点的优先级高于另一个租车点的优先级，则将与所述目标车型匹配的、第三数目的待调度车辆确定为所述目标租车点确定的调度车辆。
一种可能的实施方式中，处理器1301执行的指令中，从多个待调度车辆中，为所述目标租车点确定调度车辆，包括：
获取所述目标租车点对应的车辆需求信息；
基于所述历史车辆属性信息、所述历史租车订单对应的租用车辆属性信息和所述车辆需求信息，以及预设的待调度车辆的车辆属性信息，从多个待调度车辆中，为所述目标租车点确定调度车辆。
一种可能的实施方式中，处理器1301执行的指令中，从多个待调度车辆中，为所述目标租车点确定调度车辆，包括：
获取访问所述目标租车点的用户兴趣信息；
基于所述历史车辆属性信息、所述历史租车订单对应的租用车辆属性信 息和所述用户兴趣信息，以及预设的待调度车辆的车辆属性信息，从多个待调度车辆中，为所述目标租车点确定调度车辆。
一种可能的实施方式中，处理器1301执行的指令中，还包括：
获取所述目标租车点的回调车辆订单；所述回调车辆订单中包括回调车辆的历史服务信息；
若基于所述回调车辆的历史服务信息，确定所述回调车辆满足回调条件，则指示所述目标租车点控制所述回调车辆进行回调。
一种可能的实施方式中，处理器1301执行的指令中，指示所述目标租车点控制所述回调车辆进行回调，包括：
基于所述目标租车点的位置信息，以及预设的多个车辆回收中心的位置信息，确定所述目标租车点分别与每个所述车辆回收中心之间的行驶距离；
将最短行驶距离对应的车辆回收中心确定为接收所述回调车辆的车辆回收中心；
指示所述目标租车点将所述回调车辆调度到确定的车辆回收中心。
一种可能的实施方式中，处理器1301执行的指令中，基于所述目标租车点的位置信息，确定与所述目标租车点对应的位置区域，包括：
以所述目标租车点的位置信息为圆心、以距离阈值为半径形成的圆形区域作为与所述目标租车点对应的位置区域。
本申请实施例还提供了一种计算机可读存储介质，该计算机可读存储介质上存储有计算机程序，该计算机程序被处理器运行时执行上述车辆调配方法的步骤。
具体地，该存储介质能够为通用的存储介质，如移动磁盘、硬盘等，该存储介质上的计算机程序被运行时，能够执行上述车辆调度方法，用于解决现有技术租车点租车效率低的问题。
示例性地，可将该车辆调配方法及其硬件装置应用于租车平台的场景。目标租车点的服务器首先可以确定其目标租车区域信息，根据该目标租车区域信 息向租车平台的后端服务器发送定位信息属于该目标租车区域的用户的历史租车订单的数据获取请求，租车平台的后端服务器可以响应于该数据获取请求，向目标租车点的服务器传输相应数据；目标租车点的服务器根据接收的目标租车区域的用户的历史租车订单数据以及目标租车点被租用车辆的历史车辆属性信息，然后可以通过其预存的预测模型进行数据处理，获取调度车辆信息；根据调度车辆信息，目标租车点的服务器可以向相应中心仓的服务器发送包含车辆调度信息的相应车辆调度请求，中心仓的服务器根据车辆调度请求进行相应车辆调度操作，同时在车辆调度过程及完成后，进行目标租车点的服务器的数据更新及租车平台的数据更新。当位于该目标租车区域且有租车需求的用户通过终端登录租车平台的相应用户端租车界面时，通过查询可以获知对应于该目标租车区域的目标租车点的车型情况(车型型号、车型剩余数量、目标租车点地址等)选定车型后进入车型详情界面，可以根据提供的租车方案选择符合自己的方案去下单或是根据提供的租车方案进行目标租车点(例如相应线下门店)咨询等操作；短租类型的界面包括取车点和时间，然后点击去选车，进入选车列表等等。
本申请实施例在获取目标租车点的位置信息和该目标租车点被租用车辆的历史车辆属性信息后，确定与目标租车点对应的位置区域，获取定位信息属于该位置区域的用户的历史租车订单，基于历史车辆属性信息和历史租车订单对应的租用车辆属性信息，以及待调度车辆的车辆属性信息，从待调度车辆中，为目标租车点确定调度车辆，并将确定的调度车辆的信息发送至调度终端，从而能够将确定的车辆调度到所述目标租车点，这样，提高了目标租车点的线下访问率，也提高了目标租车点的租车效率，同时，用户能够在目标租车点直接提车，而无需去中心仓看车提车，节约了用户时间，提高了用户的体验，提高了平台的服务质量。
图14是根据本申请一些实施例所示的车辆调配方法1400的示例性流程图。在一些实施例中，方法1400可以由处理器执行。
步骤1401，获取目标租车点的关联特征数据，关联特征数据包括：目标 租车点的环境信息和目标租车区域的选品相关特征数据；
步骤1402，基于预测模型对目标租车点的环境信息和目标租车区域的选品相关特征数据进行处理，确定目标租车点的车型需求；
步骤1403，至少基于关联车辆需求，确定目标租车点的车辆调度方案。
本实施例提供的车辆调配方法，通过预测模型对目标租车点的环境信息和目标租车区域的选品相关特征进行分析，确定目标租车点的车型需求。该方法节约了在人工排序和筛选上投入的人力成本，通过预测模型确定的车型需求能够有效的贴合目标租车区域环境及当地市场的租车需求，从而提升租车效率。
图15是根据本申请一些实施例所示的车辆调配方法1500的示例性流程图。在一些实施例中，方法1500可以由处理器执行。
步骤1501，获取目标租车点的关联特征数据，关联特征数据包括：目标租车点的环境信息和目标租车区域的选品相关特征数据；
步骤1502，基于预测模型对目标租车点的环境信息和目标租车区域的选品相关特征数据进行处理，确定目标租车点的车型需求；
步骤1503，基于待调度车辆的第三车辆属性信息和目标租车点的车型需求，从待调度车辆中确定车辆调度方案的调度车辆。
图16是根据本申请一些实施例所示的车辆调配方法1600的示例性流程图。在一些实施例中，方法1600可以由处理器执行。
步骤1601，获取目标租车点的关联特征数据，关联特征数据包括：目标租车点的环境信息和目标租车区域多个维度的历史选品需求相关数据；
步骤1602，基于预测模型对所述目标租车点的环境信息和目标租车区域多个维度的历史选品需求相关数据进行处理，确定目标租车点的目标车型及其对应的目标车型数目；
步骤1603，基于待调度车辆的第三车辆属性信息和目标租车点的目标车型及其对应的目标车型数目，选定与目标租车点的目标车型匹配、目标车型数目的调度车辆。
本实施例提供的车辆调配方法，通过预测模型对目标租车点的环境信息和目标租车区域多个维度的历史选聘需求相关数据进行处理，确定目标租车点的匹配目标车型及相应的数目，并基于此选定调度车辆。该方法在节约人力成本的同时，还能兼顾市场变动、车辆供应商动态等现实因素，及时有效的贴合目标租车区域的环境及当地租车市场的需求，提升了租车效率。
本申请一些实施例提供了一种车辆调配方法及其硬件装置。在一些实施例中，目标租车点的环境信息可以选取目标城市中各个中心仓的环境信息，目标租车区域多个维度的历史选品需求相关数据包括每个中心仓自身的特征数据以及目标城市的独有特征，待调度车辆的第三车辆属性信息可以选取预设的待调度车辆的车辆属性信息，通过预测模型处理，确定目标租车点的车辆调度方案。
图17是该车辆调配方法及其硬件装置应用于用户使用租车平台的应用场景示意图。如图17所示，该场景包括：用户和租车平台；其中，用户可以在自身的终端设备上打开租车平台的租车功能，用户可以在租车界面中选择租车类型，比如，长租、短租、共享汽车。不同的租车类型，对应不同的选车场景，比如，长租类型的界面可以直接选择租车的车型，选定车型后进入车型详情界面，可以根据提供的租车方案选择符合自己的方案去下单或是根据提供的租车方案进行门店咨询等操作；短租类型的界面包括取车点和时间，然后点击去选车，进入选车列表；共享汽车类型的界面可以直接选择去选车，然后进入选车列表等。可见，租车平台上可以为用户提供多种租车方案，并且提供多种租车的车型，因此，为了提高与用户需求的匹配度，对租车平台上中心仓的选品尤为重要，精准度越高，则匹配度越高，进而提高中心仓的车辆周转率。由于现有的技术是通过人工统计数据源，然后通过人工经验决定中心仓的选品存在人工干预，导致数据以及选品的不精准，无法根据最新的市场行情以及用户需求的变动进行及时调整，并且也没有考虑地域差异带来的车型选择上的差异，比如，杭州司机开电动车的占比非常多，电动车相比油车更受欢迎；而在西北地区，由于充电桩的建设不完善以及地广人稀的特点，人们对于油车更加青睐，所以人工统计的数据不具 有各个城市以及各个中心仓的特征意义，因此，为了克服上述问题，本申请结合多方实时数据源，通过机器学习对各车型进行需求预测，由于多方因素的数据考虑到中心仓以及所在城市的信息，不仅使得预测值准确度高，也使得预测的结果更加符合用户需求，即提高了与用户需求的匹配度；然后结合预测值确定未来一定时间段内该地区对中心仓中车型车辆的需求，进而提前为中心仓调配补入，保证车辆的周准率。具体地，下述车辆调配方法旨在解决现有技术的如上技术问题。
请参见图18，示出了根据本申请一些实施例所示的车辆调配方法1800的示例性流程图。方法1800可以由处理器来实现。
步骤1801，获取目标城市在第一预设历史时间段内的特征数据。所述目标城市为设有至少一个所述中心仓的城市，所述特征数据用于表示所述目标城市中各个中心仓的环境信息以及所在目标城市的车型的相关联数据。
在一些实施例中，执行主体可以为租车平台的服务器。
在一些实施例中，服务器获取某个城市中在一个预设历史时间段内的特征数据，比如，该城市在一个月前所对应的特征数据。为了方便理解，以下将该城市视为目标城市，这里的目标城市设有至少一个中心仓，中心仓用于存放各种车型的车辆，中心仓的车辆是动态的来源于各个租赁公司的。
其中，目标城市对应的特征数据中包括每个中心仓自身的特征数据以及目标城市的独有特征。中心仓自身的特征数据可以包括中心仓所在位置的环境信息，比如中心仓所能够覆盖目标城市中区域的人口密集程度、道路特征，如交叉口数量、形态、交通是否便利等。目标城市的独有特征可以包括整个城市的客观环境的信息、客观环境的信息相关的车辆信息，比如整个城市的人口密集程度、道路特征、与车型相关信息(比如，平台上下单的车型、该城市较为偏好的车型等)。因此，目标城市对应的特征数据具有该城市或者各个所选中心仓的特征，该特征数据具有的多方影响因素为中心仓选品的精准度提供了有效地数据源。
步骤1802，通过预测模型对所述特征数据进行处理，得到所述目标城市中各个中心仓在预设时间段内对应的车型需求预测数据。所述预测模型是由决 策树模型训练得到。
本实施例中，由于各个中心仓选品直接影响到中心仓的车辆周转效率，为了提高周转效率，可以通过预测目标城市在未来一个时间段对车型的需求量，进而及时调整中心仓的车型以及数量。
具体地，为了使得预测的结果准确度高，采用决策树算法，通过历史特征数据对决策树模型进行训练得到预测模型，然后对目标城市当前的特征数据，比如前一个月的特征数据，通过数据预处理后输入到预测模型中，得到该目标城市中每个中心仓在预设时间段内(比如未来一周时间)内对车型需求的预测值即车型需求预测数据。虽然预测的是未来一段时间内的车型需求量比如未来一周的车型需求量，但是在当前时间节点的下一个正常日，依然可以重新获取以前的历史数据(包括当前时间节点的数据，比如一个月的数据)来预测下一个未来一周的车型需求量，因此，该预测的方式可以是日更新的动态迭代的，实现了中心仓的精细化选品。
由于该预测方式的输入数据是考虑到多方因素的特征数据，更加符合所在地域居民的需求，因此，该车型需求预测数据的确定，能够为中心仓选品与用户需求产生较高的匹配度。
可选的，例如在图18所述的实施例的基础上，对步骤1802进行了详细说明。所述根据所述特征数据，通过预测模型，得到所述目标城市中各个中心仓在预设时间段内对应的车型需求预测数据，包括：
步骤a1，对所述特征数据进行归一化处理，得到所述特征数据分布在所述目标城市中各个中心仓对应的目标特征量；
步骤a2，将所述目标特征数据对应的特征量输入到所述预测模型中，得到所述目标城市中各个中心仓在预设时间段内对车型需求的车型需求预测数据。
在一些实施例中，首先对目标城市的特征数据进行归一化处理，即进行特征量化，转换成数字特征。比如，将语言特征转化为数字型向量。然后对目标城市的特征数据整合成至少一条数据，每条数据包括一个中心仓对应的特征量和 该目标城市对应的固有特征量(如上述的整个城市的人口密集程度、道路特征、与车型相关信息等)，然后将每条数据输入到预测模型中，能够得到该目标城市中每个中心仓在预设时间段内对车型需求的车型需求预测数据。该预测结果融合了目标城市中每个中心仓的特征和该目标城市的道路、人口以及对车型的喜好等因素，因此，该车型需求预测数据的确定，不但数据精准，还使得中心仓停放的车辆能够和用户的需求匹配度高，进而提高了车辆周转率。
步骤1803，根据所述车型需求预测数据，确定各个所述中心仓的选品结果。
在一些实施例中，由于根据各个中心仓的环境信息，如中心仓所在的地理位置以及规模大小，可以确定出该每个中心仓所能覆盖的用户区域，因此，可以将每个中心仓对应的车型需求预测数据作为选品的依据，比如将该中心仓所能覆盖的用户区域的车型按预测数据大小排序，最终选择排序前几的车型作为该中心仓的选品结果。
上述实施例提供的车辆调配方法，通过获取的目标城市中各个中心仓的环境信息以及所在目标城市的车型的相关联数据，考虑多方因素，实现了全链路数据的打通，再结合机器学习算法，比如通过训练决策树模型实现了车型需求预测，无需人工干预，也大大节省了人力资源，然后基于预测数据，对中心仓的选品进行决策，实现了精准地选品，进而使得中心仓停放的车辆与用户的需求匹配度高，使得中心仓停放的车辆能够和用户的需求匹配度高，提高了车辆周转率。
可选的，比如在图18所述的实施例的基础上，对步骤1803进行了详细说明。其中，所述车型需求预测数据包括多个预选车型的预测需求量。所述根据所述车型需求预测数据，确定各个所述中心仓的选品结果，包括：
步骤b1，根据所述多个预选车型中每个预选车型的预测需求量，通过需求转化率，计算所述目标城市中各个中心仓在预设时间段内对每个预选车型需求的第一实际需求量；
步骤b2，针对每个中心仓，对每个所述预选车型对应的第一实际需求量 依据从高到低的顺序进行排序；
步骤b3，将在预设排名顺序内的所有预选车型作为所述中心仓的选品结果。
在一些实施例中，每个中心仓对应的车型需求预测数据中包含了多个预选车型以及每个预选车型的需求量，这里预测的未来预设时间段(比如，未来一周时间)内对每个预选车型的需求量是较为符合该中心仓所在目标城市覆盖区域的需求，即与用户需求具有较高的匹配度，且由于中心仓所在地域环境以及规模限制，需要从多个预选车型中选择最终的车型。因此，可以依据每个中心仓对应的车型需求预测数据中预选车型的需求量大小，来确定最终选择的车型。
具体地，首先将每个预选车型的预测需求量转化成第一实际需求量，比如，通过需求转化率来计算：每个预选车型的预测需求量*需求转化率＝每个预选车型的第一实际需求量。
可选地，需求转化率是由历史需求量和所述历史需求量对应的预测值作比值得到的；所述历史需求量对应的预测值是对所述决策树模型训练过程中得到的。
然后针对每个中心仓，基于每个预选车型的第一实际需求量，将所有预选车型依据数量从高到低的顺序排序，将在预设排名顺序内的所有预选车型作为所述中心仓的选品结果，比如，将顺序的前十名预选车型作为所述中心仓的选品车型，且选品结果中包含每个选定的车型的需求量。这样就可以通过选定的车型以及需求量对中心仓中当前的车型情况进行合理补充或是合理调度。
为了实现预测，需要建立预测模型，参见图19所示，图19是根据本申请实施例所示的车辆调配方法1900的示例性流程图，本实施例在上述实施例的基础上，例如在图18所述的实施例的基础上，对如何建立预测模型进行了详细说明。在所述得到所述目标城市中各个中心仓在预设时间段内对应的车型需求预测数据之前，所述方法还包括：
步骤1901，获取多个预定城市中的每个预定城市在第二预设历史时间段 内的历史特征数据和每个所述预定城市的各个中心仓在第三预设历史时间段内对各个车型的历史需求量，所述历史特征数据包括多个维度的数据；
步骤1902，根据每个所述预定城市对应的所述多个维度的数据和所述历史需求量，对所述决策树模型进行训练，得到所述预测模型。
其中，第一预设历史时间段的时间间隔等于第二预设历史时间段的时间间隔，第二预设历史时间段的时间间隔大于第三预设历史时间段的时间间隔，预设时间段的时间间隔等于第三预设历史时间段的时间间隔。
在实际应用中，第一预设历史时间段可以为当前时间之前的一个月，比如2019.10-2019.11，时间间隔为一个月；第二预设历史时间段可以为一个也可以为多个，可以与第一预设历史时间段上的时间有交叉，第二预设历史时间段得终止时间也可以在第一预设历史时间段的起始时间之前(比如，2019.9-2019.10)，时间间隔为一个月；第三预设历史时间段的起始时间可以为第二预设历史时间段的终止时间(比如2019.10.1-2019.10.7)，第三预设历史时间段的的时间间隔为7天即一周时间，预设时间段的起始时间可以为第一预设历史时间段的终止时间(比如，2019.11.1-2019.11.7)，预设时间段的时间间隔为7天。
具体地，首先选取多个城市即多个预定城市，每个预设城市设有一个或多个中心仓，多个预定城市中可以包括目标城市也可以不包括，在此不做限定。针对每个预定城市，获取每个预定城市在当前时间之前的某一个月或几个月中每个月对应的历史特征数据，这样第二预设历史时间段可以为一个或多个，然后针对每个月对应的历史特征数据进行预处理。其中，每个预定城市在一个第二预设历史时间段内的历史特征数据包括该预定城市的固有特征数据以及该预定城市中各个中心仓的特征数据，因此，一个第二预设历史时间段内的历史特征数据可以包括多个维度的数据。然后根据以多个维度的数据和该预定城市中各个中心仓对应的历史需求量作为训练数据，对决策树模型进行训练，得到预测模型。
其中，多个维度可以包括：租车平台下单维度、租车平台冒泡维度、租车平台司机维度、城市出行维度、中心仓环境维度、城市人口密度维度、城市环境 维度、城市租赁公司维度以及城市人口收入维度。
具体地，租车平台下单维度的数据是表示中心仓下单车型的数量，租车平台冒泡维度的数据是表示用户在租车平台上选择过的车型但并未下单的数量，租车平台司机维度的数据是表示该租车平台上在该预定城市的司机数量，城市出行维度的数据是表示该预定城市所有人口的月均历史出行数据量，中心仓环境维度的数据是表示中心仓所覆盖的区域的人口密集程度和中心仓周围道路特征，城市人口密度维度的数据是表示该预定城市的人口密集程度，城市环境维度的数据是表示该预定城市的道路特征，城市租赁公司维度的数据是表示该预定城市中所有租赁公司口碑以及规模，城市人口收入维度的数据是表示该预定城市所有人口的月均收入。
因此，这些维度考虑了所在城市租车平台上的下单车型数量和用户在租车平台上的行为数据(比如用户在租车平台上选择过的车型但并未下单的数量)，还考虑了所在城市的人口密度、交通环境、收入，也考虑了具体到每个中心仓所覆盖区域的人口密度、交通环境，还考虑了选取车型入中心仓的租赁公司的口碑以及规模等等可能引起车型需求预测数据准确度的多方因素，将多方因素的数据作为训练数据对决策树的训练提高了参数的精度，使得训练后得到的预测模型更优，进而保证了预测结果的准确度。
具体地，参见图20，图20是根据本申请一些实施例所示的车辆调配方法2000的流程示意图。如何基于每个所述预定城市对应的所述多个维度的数据和所述历史需求量，对所述决策树模型进行训练，得到所述预测模型，可以通过如下步骤实现：
步骤2001，对所述多个维度的数据进行归一化处理，得到多个维度的特征量。
可选地，对所述中心仓环境维度的数据、所述城市环境维度的数据以及所述城市租赁公司维度的数据进行预定义编码，得到所述中心仓环境维度的特征量、所述城市环境维度的特征量以及所述城市租赁公司维度的特征量。其中，一 个中心仓对应一个中心仓环境维度的特征量。
将所述租车平台下单维度的下单车型数量作为所述租车平台下单维度的特征量，将所述租车平台冒泡维度的用户冒泡数量作为所述租车平台冒泡维度的特征量，将所述租车平台司机维度的司机数量作为所述租车平台司机维度的特征量，将所述城市出行维度的月均出行数量作为所述城市出行维度的特征量，将所述城市人口密度维度的人口密度值作为所述城市人口密度维度的特征量，将所述城市人口收入维度的人均月收入值作为所述城市人口收入维度的特征量。
步骤2002，以每个所述预定城市的每个中心仓对应的多个维度的特征量和所述历史需求量为一个训练样本，对所述决策树模型进行训练，其中，所述历史需求量为所述决策树模型训练过程中的标签；
步骤2003，根据所述决策树模型的输出，以及作为所述标签的所述历史需求量之间的差异，调整所述决策树模型的参数，直至所述决策树模型达到期望训练效果；
步骤2004，将所述达到期望训练效果的决策树模型作为所述预测模型。
本实施例中，若每个预定城市设有多个中心仓，则每个预定城市的多个维度的特征以及多个中心仓中的每个中心仓对应的历史需求量可以分为多个样本，其中样本的个数与该预定城市中中心仓的个数一致。
例如，预定城市1设有2个中心仓(中心仓11和中心仓12)，则该预定城市1的多个维度的特征以及每个中心仓对应的历史需求量可以分为2个样本(样本11和样本12)，则样本11包括中心仓11环境维度的特征量、城市环境维度的特征量、城市租赁公司维度的特征量、租车平台下单维度的特征量、租车平台冒泡维度的特征量、租车平台司机维度的特征量、城市出行维度的特征量、城市人口密度维度的特征量、城市人口收入维度的特征量、中心仓11对应的历史需求量；样本12包括中心仓12环境维度的特征量、城市环境维度的特征量、城市租赁公司维度的特征量、租车平台下单维度的特征量、租车平台冒泡维度的特征量、租车平台司机维度的特征量、城市出行维度的特征量、城市人口密度维 度的特征量、城市人口收入维度的特征量、中心仓12对应的历史需求量，因此，每个预定城市的每个样本中包含的各个中心仓环境维度的特征量不同，但是有关各个中心仓所在同一城市的其他维度的特征量相同。因此，这样的训练样本不仅仅具有用户的地域特性还有中心仓特性，使得为各个中心仓的选品具有针对性和唯一性，进而使得中心仓的车型与用户的需求匹配度高。
具体地，训练样本为模型的输入量，将历史需求量为所述决策树模型训练过程中的标签，输出量为车型需求预测数据，然后将输出量与标签进行比对，计算误差，然后对决策树模型的参数进行反馈调整，直到输出量与标签的误差较小且趋于稳定，即决策树模型达到期望训练效果，将该训练后的决策树模型作为预测模型。
可选的，如何通过训练样本对决策树模型进行训练，本实施例在上述实施例的基础上，例如，在图20所述的实施例的基础上，对步骤2002进行了详细说明。所述以每个所述预定城市的每个中心仓对应的多个维度的特征量和所述历史需求量为一个训练样本，对所述决策树模型进行训练，包括以下步骤：
步骤c1、根据每个所述预定城市的每个中心仓对应的各个所述维度的特征量，生成第一矩阵；
步骤c2、根据每个所述预定城市的每个中心仓对应的各个车型的历史需求量，生成第二矩阵；
步骤c3、根据所述第一矩阵和所述第二矩阵，形成所述训练样本，所述训练样本为所述第一矩阵和所述第二矩阵的合并矩阵，其中，所述第一矩阵为所述合并矩阵中的第一输入量X，所述第二矩阵作为所述合并矩阵的标签输入量Y，所述第一输入量X对应唯一的所述标签输入量Y；
步骤c4、将所述第一输入量X和标签输入量Y同步输入到所述决策树模型中训练，输出所述标签输入量Y对应的预测值。
在一些实施例中，每个中心仓对应的各个所述维度的特征量包括该中心仓环境维度的特征量、城市环境维度的特征量、城市租赁公司维度的特征量、租 车平台下单维度的特征量、租车平台冒泡维度的特征量、租车平台司机维度的特征量、城市出行维度的特征量、城市人口密度维度的特征量、城市人口收入维度的特征量，则每个中心仓对应的各个所述维度的特征量生成的第一矩阵为[该中心仓环境维度的特征量城市环境维度的特征量城市租赁公司维度的特征量租车平台下单维度的特征量租车平台冒泡维度的特征量租车平台司机维度的特征量城市出行维度的特征量城市人口密度维度的特征量城市人口收入维度的特征量]，第二矩阵是由中心仓对应的各个车型的历史需求量生成的，即[车型1的历史需求量车型2的历史需求量…车型N的历史需求量]，则训练样本为[该中心仓环境维度的特征量城市环境维度的特征量城市租赁公司维度的特征量租车平台下单维度的特征量租车平台冒泡维度的特征量租车平台司机维度的特征量城市出行维度的特征量城市人口密度维度的特征量城市人口收入维度的特征量车型1的历史需求量车型2的历史需求量…车型N的历史需求量]，针对每个训练样本的输出结果为每个车型的历史需求量对应的预测值或是该中心仓对每个车型的车型需求预测数据。
可选的，为了使得车辆调配业务中中心仓的选品结果能够提高用户的满意度，参见图21所示，图21是根据本申请一些实施例所示的车辆调配方法2100的示例性流程图，本实施例在上述实施例的基础上，例如在图21所述的实施例的基础上，对中心仓选品方法进行了详细说明。在所述确定各个所述中心仓的选品结果之后，所述方法还可以包括：
步骤2101，将所述选品结果对应的车型推送至各个用户端，以使各个所述用户端的用户针对所述选品结果对应的车型进行意见反馈；
步骤2102，接收各个所述用户端的反馈信息，并根据所述反馈信息，调整所述选品结果。
图22是根据本申请一些实施例所示的车辆调配方法2100的应用场景示例图。
本实施例中，服务器可以将计算出的选品结果即所需车型通过租车平台 推送给各个用户端，各个用户端接收到推送消息后，根据推送消息中的提示对推送的车型进行意见反馈。比如，参见图22所示，租车平台的推送消息为：“请对以下车型进行满意度打分：5分、3分、1分”，用户通过用户端根据自己喜好选择对应分值，然后点击“提交”，用户端将用户提交的结果反馈至租车平台中的服务器，服务器根据反馈意见适应调整对选品结果中各个车型的需求量进行调整，使得调整后的选品结果能够使得用户有较高的满意度。
可选的，为了及时的填补中心仓的车型需求的空缺，以及及时向租赁公司进行协商，进而提高中心仓中车辆的周转率，在所述确定各个所述中心仓的选品结果之后，所述车辆调配方法还包括：
根据各个所述中心仓的选品结果，以及各个所述中心仓当前的各个车型数量，计算各个所述中心仓的待补入的各个车型数量；或者，根据各个所述中心仓对应的调整后的选品结果，以及各个所述中心仓当前的各个车型数量，计算各个所述中心仓的待补入的各个车型数量。
其中，这两种计算方式均为计算中心仓当前的车型的数据与选品结果中对应车型的需求量之差，将差值作为中心仓的待补入的各个车型数。除此之外，还可以通过其他算法来补入该中心仓中各个车型数量，保证中心仓中各种车型车辆的后续调度。
可选地，训练样本存储在样本库中；为了动态更新样本或是优化预测模型，在所述确定各个所述中心仓的选品结果之后，所述方法还包括：
获取所述目标城市中各个中心仓在预设时间段内对车型需求的第二实际需求量；根据所述特征数据和所述第二实际需求量，更新所述样本库。
在一些实施例中，为了在实际应用中不断地优化预测模型，可以在经过上述预设时间段后，获取目标城市中各个中心仓在该预设时间段内对车型需求的真实值即第二实际需求量，然后将该真实值替换之前通过预测模型预测得到的第一实际需求量，将目标城市中每个中心仓对应的多维的特征量和对应的所述第二实际需求量作为新的样本存储至数据库中，使得数据库能够动态更新。
可选地，在得到选品结果后还可以更新需求转换率，进而提高选品结果的准确度，即在所述确定各个所述中心仓的选品结果之后，该方法还可以包括：
获取所述目标城市中各个中心仓在预设时间段内对车型需求的第二实际需求量；根据所述第一实际需求量和所述第二实际需求量，通过误差计算，调整所述需求转化率。
本实施例中，为了适时调整需求转化率，可以在经过上述预设时间段后，获取目标城市中各个中心仓在该预设时间段内对车型需求的真实值即第二实际需求量，然后将真实值与之前通过预测模型预测得到的第一实际需求量进行误差计算，进而达到调整需求转化率的目的，保证预测结果经过需求转化率转换后的选品结果精准度高。
上述实施例提供的车辆调配方法，在整个过程中，可以是日更新的动态迭代的，每天根据中心仓已停放的车辆信息以及计算得到的该中心仓的选品结果进行比较，自动化的决策出还需要让车辆提供方(比如租赁公司)为提供哪种车型的车辆以及各提供多少辆，该方案实现了中心仓的精细化选品，并且是日更新的，通过全链路数据的打通，实现了端到端的预测和决策，无需人工干预，大大节省了运营人力，并且能够更加精准的选品，使得中心仓停放的车辆能够和用户的需求匹配度高，提高了车辆周转率。
图23为根据本申请一些实施例所示的车辆调配装置2300的示例性结构图。该车辆调配装置具体可以是上述实施例中的租车平台。该车辆调配装置可以执行车辆调配方法实施例提供的处理流程，如图23所示，车辆调配装置2300包括：第三获取模块2301、预测模块2302和选品模块2303；其中，第三获取模块2301，用于获取目标城市在第一预设历史时间段内的特征数据，所述目标城市为设有至少一个所述中心仓的城市，所述特征数据用于表示所述目标城市中各个中心仓的环境信息以及所在目标城市的车型的相关联数据；预测模块2302，用于根据所述特征数据，通过预测模型，得到所述目标城市中各个中心仓在预设时间段内对应的车型需求预测数据，所述预测模型是由决策树模型训练得到；选品 模块2303，用于根据所述车型需求预测数据，确定各个所述中心仓的选品结果。
如前所述，获取模块501可以用于获取目标城市在第一预设历史时间段内的特征数据，所述目标城市为设有至少一个所述中心仓的城市，所述特征数据用于表示所述目标城市中各个中心仓的环境信息以及所在目标城市的车型的相关联数据。在一些实施例中，获取模块501可以包括第三获取模块2301。
在一些实施例中，前述的第一确定模块502可以包括预测模块2302。在一些实施例中，前述的第二确定模块503可以包括选品模块2303。
可选的，所述装置2300还包括：模型建立模块2304，模型建立模块2304包括：第一获取单元和模型训练单元。
第一获取单元，用于获取多个预定城市中的每个预定城市在第二预设历史时间段内的历史特征数据和每个所述预定城市的各个中心仓在第三预设历史时间段内对各个车型的历史需求量，所述历史特征数据包括多个维度的数据。
模型确定单元，用于根据每个所述预定城市对应的所述多个维度的数据和所述历史需求量，对所述决策树模型进行训练，得到所述预测模型；其中，第一预设历史时间段的时间间隔等于第二预设历史时间段的时间间隔，第二预设历史时间段的时间间隔大于第三预设历史时间段的时间间隔，预设时间段的时间间隔等于第三预设历史时间段的时间间隔。
可选的，所述多个维度包括：租车平台下单维度、租车平台冒泡维度、租车平台司机维度、城市出行维度、中心仓环境维度、城市人口密度维度、城市环境维度、城市租赁公司维度以及城市人口收入维度；所述模型确定单元，包括：数据处理子单元、模型训练子单元、参数调整子单元和模型确定子单元。
数据处理子单元，用于对所述多个维度的数据进行归一化处理，得到多个维度的特征量。
模型训练子单元，用于以每个所述预定城市的每个中心仓对应的多个维度的特征量和所述历史需求量为一个训练样本，对所述决策树模型进行训练，其中，所述历史需求量为所述决策树模型训练过程中的标签。
参数调整子单元，用于根据所述决策树模型的输出，以及作为所述标签的所述历史需求量之间的差异，调整所述决策树模型的参数，直至所述决策树模型达到期望训练效果。
模型确定子单元，用于将所述达到期望训练效果的决策树模型作为所述预测模型。
可选的，数据处理子单元，具体用于：对所述中心仓环境维度的数据、所述城市环境维度的数据以及所述城市租赁公司维度的数据进行预定义编码，得到所述中心仓环境维度的特征量、所述城市环境维度的特征量以及所述城市租赁公司维度的特征量；将所述租车平台下单维度的下单车型数量作为所述租车平台下单维度的特征量，将所述租车平台冒泡维度的用户冒泡数量作为所述租车平台冒泡维度的特征量，将所述租车平台司机维度的司机数量作为所述租车平台司机维度的特征量，将所述城市出行维度的月均出行数量作为所述城市出行维度的特征量，将所述城市人口密度维度的人口密度值作为所述城市人口密度维度的特征量，将所述城市人口收入维度的人均月收入值作为所述城市人口收入维度的特征量。
可选的，模型训练子单元，具体用于：根据每个所述预定城市的每个中心仓对应的各个所述维度的特征量，生成第一矩阵；根据每个所述预定城市的每个中心仓对应的各个车型的历史需求量，生成第二矩阵；根据所述第一矩阵和所述第二矩阵，形成所述训练样本，所述训练样本为所述第一矩阵和所述第二矩阵的合并矩阵，其中，所述第一矩阵为所述合并矩阵中的第一输入量X，所述第二矩阵作为所述合并矩阵的标签输入量Y，所述第一输入量X对应唯一的所述标签输入量Y；将所述第一输入量X和标签输入量Y同步输入到所述决策树模型中训练，输出所述标签输入量Y对应的预测值。
可选的，所述预测模块，具体用于：对所述特征数据进行归一化处理，得到所述特征数据分布在所述目标城市中各个中心仓对应的目标特征量；将所述目标特征数据对应的特征量输入到所述预测模型中，得到所述目标城市中各个 中心仓在预设时间段内对车型需求的车型需求预测数据。
可选的，所述车型需求预测数据包括多个预选车型的预测需求量；所述选品模块，具体用于：根据所述多个预选车型中每个预选车型的预测需求量，通过需求转化率，计算所述目标城市中各个中心仓在预设时间段内对每个预选车型需求的第一实际需求量；针对每个中心仓，对每个所述预选车型对应的第一实际需求量依据从高到低的顺序进行排序；将在预设排名顺序内的所有预选车型作为所述中心仓的选品结果。
可选的，所述需求转化率是由历史需求量和所述历史需求量对应的预测值作比值得到的；所述历史需求量对应的预测值是对所述决策树模型训练过程中得到的。
可选的，所述装置还包括：第一调整模块2305；第一调整模块，用于在所述确定各个所述中心仓的选品结果之后，将所述选品结果对应的车型推送至各个用户端，以使各个所述用户端的用户针对所述选品结果对应的车型进行意见反馈；接收各个所述用户端的反馈信息，并根据所述反馈信息，调整所述选品结果。
可选的，所述装置还包括：第一待补入数据确定模块2306；第一待补入数据确定模块2306，用于在所述确定各个所述中心仓的选品结果之后，根据各个所述中心仓的选品结果，以及各个所述中心仓当前的各个车型数量，计算各个所述中心仓的待补入的各个车型数量。
可选的，所述装置还包括：第二待补入数据确定模块2307；第二待补入数据确定模块2307，用于在所述调整所述选品结果之后，根据各个所述中心仓对应的调整后的选品结果，以及各个所述中心仓当前的各个车型数量，计算各个所述中心仓的待补入的各个车型数量。
可选的，所述训练样本存储在样本库中；所述装置还包括：第一更新模块2108；更新模块2308，用于在所述确定各个所述中心仓的选品结果之后，获取所述目标城市中各个中心仓在预设时间段内对车型需求的第二实际需求量；根 据所述特征数据和所述第二实际需求量，更新所述样本库。
可选的，所述装置还包括：第二调整模块2309；第二调整模块2309，用于在所述确定各个所述中心仓的选品结果之后，获取所述目标城市中各个中心仓在预设时间段内对车型需求的第二实际需求量；根据所述第一实际需求量和所述第二实际需求量，通过误差计算，调整所述需求转化率。
图23所示实施例的车辆调配装置可用于执行上述方法实施例的技术方案，其实现原理和技术效果类似，此处不再赘述。
上述实施例通过获取目标城市在第一预设历史时间段内的特征数据，其中目标城市为设有至少一个所述中心仓，该特征数据用于表示所述目标城市中各个中心仓的环境信息以及所在目标城市的车型的相关联数据；然后根据特征数据，通过由决策树模型训练得到的预测模型，得到所述目标城市中各个中心仓在预设时间段内对应的车型需求预测数据；再根据所述车型需求预测数据，确定各个所述中心仓的选品结果，因此，通过获取的目标城市中各个中心仓的环境信息以及所在目标城市的车型的相关联数据，考虑多方因素，实现了全链路数据的打通，再结合机器学习算法，比如通过训练决策树模型实现了车型需求预测，无需人工干预，也大大节省了人力资源，然后基于预测数据，对中心仓的选品进行决策，实现了精准地选品，进而使得中心仓停放的车辆与用户的需求匹配度高，使得中心仓停放的车辆能够和用户的需求匹配度高，提高了车辆周转率。
在整个过程中，可以是日更新的动态迭代的，每天根据中心仓已停放的车辆信息以及计算得到的该中心仓的选品结果进行比较，自动化的决策出还需要让车辆提供方(比如租赁公司)为提供哪种车型的车辆以及各提供多少辆，该方案实现了中心仓的精细化选品，并且是日更新的，通过全链路数据的打通，实现了端到端的预测和决策，无需人工干预，大大节省了运营人力，并且能够更加精准的选品，使得中心仓停放的车辆能够和用户的需求匹配度高，提高了车辆周转率。
图24是根据本申请一些实施例所示的车辆调配设备2400的示例性结构 图。该车辆调配设备具体可以是上述实施例中的租车平台。该车辆调配设备可以执行车辆调配设备方法实施例提供的处理流程，如图24所示，本实施例提供的设备2400包括：至少一个处理器2401和存储器2402。其中，处理器2401、存储器2402通过总线2403连接。
在具体实现过程中，至少一个处理器2401执行所述存储器2402存储的计算机执行指令，使得至少一个处理器2401执行上述方法实施例中的方法。
处理器2401的具体实现过程可参见上述方法实施例，其实现原理和技术效果类似，此处不再赘述。
在上述的图24所示的实施例中，应理解，处理器可以是中央处理单元(英文：Central Processing Unit，简称：CPU)，还可以是其他通用处理器、数字信号处理器(英文：Digital Signal Processor，简称：DSP)、专用集成电路(英文：Application Specific Integrated Circuit，简称：ASIC)等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。结合发明所公开的方法的步骤可以直接体现为硬件处理器执行完成，或者用处理器中的硬件及软件模块组合执行完成。
存储器可能包含高速RAM存储器，也可能还包括非易失性存储NVM，例如至少一个磁盘存储器。
总线可以是工业标准体系结构(Industry Standard Architecture，ISA)总线、外部设备互连(Peripheral Component Interconnect，PCI)总线或扩展工业标准体系结构(Extended Industry Standard Architecture，EISA)总线等。总线可以分为地址总线、数据总线、控制总线等。为便于表示，本申请附图中的总线并不限定仅有一根总线或一种类型的总线。
另外，在一些实施例中，还提供一种计算机可读存储介质，其上存储有计算机程序，所述计算机程序被处理器执行以实现上述实施例所述的车辆调配方法。
图25是根据本申请一些实施例所示的车辆调配方法2500的示例性流程 图。在一些实施例中，方法2500可以由处理器执行。
步骤2501，获取目标租车点的关联特征数据，关联数据特征包括：目标租车点的历史周转车辆信息及相关信息和目标租车区域的库位信息。
步骤2502，基于预测模型对目标租车点的历史周转车辆信息及相关信息和目标租车区域的库位信息进行处理，确定目标租车点的车辆数需求。
步骤2503，至少基于目标租车点的车辆数需求，确定目标租车点的车辆调度方案。
通过预测模型对目标租车点的历史周转车辆信息及相关信息和目标租车区域的库位信息进行处理，减少时效性、库位分配数量不合理、不灵活、效率低、人工成本高等缺陷，可以更高效地预测确定目标租车点在目标租车区域的待调配车辆数需求，从而实现更灵活合理有效的车辆调配方案，最终提高租车业务效率，满足用户需求，提高用户体验。
图26是根据本申请一些实施例所示的车辆调配方法2600的示例性流程图。在一些实施例中，方法2600可以由处理器执行。
步骤2601，获取目标租车点的关联特征数据，关联数据特征包括：目标租车点的历史周转车辆信息及相关信息和目标租车区域的库位信息。
步骤2602，基于预测模型对目标租车点的历史周转车辆信息及相关信息和目标租车区域的库位信息进行处理，确定目标租车点的车辆数需求。
步骤2603，基于待调度车辆的第三属性信息和目标租车点的车辆数需求，从待调度车辆中确定车辆调度方案的调度车辆。
图27是根据本申请一些实施例所示的车辆调配方法2700的示例性流程图。在一些实施例中，方法2700可以由处理器执行。
步骤2701，获取目标租车点的关联特征数据，关联数据特征包括：目标租车点的历史周转车辆数目及多维度车辆周转影响数据和目标租车区域的库位缺口数目。
步骤2702，基于预测模型对目标租车点的历史周转车辆数目及多维度车 辆周转影响数据、目标租车区域的库位缺口数目进行处理，确定目标租车点对应目标租车区域的车辆数配额需求。
步骤2703，基于待调度车辆的第三车辆属性信息和目标租车点的车辆数配额需求，从待调度车辆中确定车辆调度方案的调度车辆。
关联特征数据选取目标租车点的历史周转车辆数目及多维度车辆周转影响数据，从多个角度更全面地采集反映目标租车点车辆周转情况数据，使得通过预测模型预测计算后，获取更贴合目标租车点实际的调配车辆需求，然后根据目标租车点的车辆数配额需求从待调度车辆确定调度车辆，提高了车辆调配效率，降低了人工统计分配的运营成本，从而促进租车业务的高效运行，满足用户更及时灵活的需求，提高用户体验。
本申请一些实施例提供了一种车辆调配方法及其硬件装置。在一些实施例中，目标租车点的关联特征数据可以选取目标租车点的历史周转车辆数目及多维度车辆周转影响数据和目标租车区域的库位缺口数目，待调度车辆的第三车辆属性信息可以选取预设的待调度车辆的车辆属性信息，通过预测模型处理，确定目标租车点的车辆调度方案。
参见图28，示出了本申请一些实施例提供的一种车辆调配方法2800的示例性流程图，方法2800可以由处理器来实现。
步骤2801，获取租赁公司在每个中心仓的缺口车辆数作为每个中心仓的初始库位配额；
步骤2802，对于每个中心仓，获取该中心仓历史时间段内的真实交付峰值，根据所述初始库位配额计算相同历史时间段内该中心仓的预测交付峰值，根据所述真实交付峰值与所述预测交付峰值的差值，通过预测模型对所述初始库位配额进行修正得到该中心仓的修正后库位配额；
步骤2803，向所述租赁公司的每个中心仓分配对应的所述修正后库位配额的车辆。
具体来说，本车辆调配方法主要应用于租车平台的服务器，由服务器为每 个租赁公司作出分配决策。租赁公司一般拥有多个中心仓，步骤2801获取租赁公司在每个中心仓的缺口车辆数。该缺口车辆数可以通过预测确定，也可以通过传感器监控，也可以通过工作人员上传数据。该缺口车辆数用于表示需要为该租赁公司新增或减少多少个库位。然后步骤2802对每个中心仓的初始库位配额进行修正，使得库位配额更为符合真实需求。最后，步骤2803向所述租赁公司的每个中心仓分配对应的所述修正后库位配额的车辆。分配方式可以通过下达工单等方式实现。
本车辆调配方法实现了租车业务中心仓的库位智能化管理，能够动态的为每个租赁公司分配库位，并且支持中心仓数量和租赁数量的增加以及每个中心仓的总库位的变动，进而大规模的扩展到其他城市，只要提供仓库的一些基础信息和订单数据即可快速拓展到其他城市。其次，该方法极大的节省了运营人力，并且提高了车位利用率和车辆周转率，充分满足租赁的需求，不会发生因为库位分配不公或者不够导致租赁没能及时满足用户的订车需求，使得平台与租赁的合作更加高效。最后，根据真实交付峰值对库位配额进行修正，使得库位配额更为符合真实需求。
参见图29，示出了本申请一些实施例提供的一种车辆调配方法2900的示例性流程图，方法2900可以由处理器实现。
步骤2901，获取租赁公司的历史租车数量、租赁公司的信息维度数据、时间数据，输入预测模型，获得预测模型输出的租赁公司的总租车数量；
步骤2902，获取租赁公司的历史还车数量、租赁公司的信息维度数据、时间数据，输入预测模型，获得预测模型输出的租赁公司的总还车数量；
步骤2903，获取租赁公司每个中心仓的历史占比，确定每个中心仓的租车数量为总租车数量×该中心仓的历史占比，每个中心仓的还车数量为总还车数量×该中心仓的历史占比；
步骤2904，获取每个中心仓的停车数量；
步骤2905，基于每个中心仓的租车数量、还车数量、以及停车数量，确 定每个中心仓的缺口车辆数作为每个中心仓的初始库位配额；
步骤2906，对于每个中心仓，获取该中心仓历史时间段内的真实交付峰值，根据所述初始库位配额计算相同历史时间段内该中心仓的预测交付峰值，所述预测交付峰值y＝X×N/A，其中，y为预测交付峰值，X为初始库位配额，N为历史时间段长度，A为车位流速；
步骤2907，如果预测交付峰值减去真实交付峰值的差值大于差值阈值，则以所述初始库位配额作为所述修正后库位配额；
步骤2908，如果预测交付峰值减去真实交付峰值的差值小于等于差值阈值，则计算所述修正后库位配额为：(真实交付峰值+差值阈值)/(N/A)；
步骤2909，向所述租赁公司的每个中心仓分配对应的所述修正后库位配额的车辆。
具体来说，步骤2901和步骤2902通过预测模型，对租赁公司的总租车数量和总还车数量进行预测。优选地，通过输入历史租车单量、租赁公司规模数据、节假日信息、天气、租赁公司信用分、车型丰富度、租赁品牌知名度等维度特征到预测模型中，做租车单量的回归，输出每个租赁公司在未来X个小时内出租车辆的预测数据。其中，X为可变参数。还车数量预测同上，通过历史老用户租期到后还车的单量数据，和租赁公司的一些特征及上下文信息，将其数据处理后输入到预测模型中进行还车数量的预测，通过模型预测输出每个租赁公司在未来X个小时内还车数量的预测数据。
然后，步骤2903通过历史占比，将总租车数量和总还车数量按照每个中心仓的历史占比，确定每个中心仓的租车数量和还车数量。再结合步骤2904获得的每个租赁公司在中心仓现有停放的车辆数，即可计算出每家租赁公司每个中心仓的缺口车辆数。此时，步骤2905可计算出还需为每个租赁公司的每个中心仓新增或减少多少个库位，作为该中心仓的初始库位配额。
当业务规模小时，若历史数据量级较小时，模型预测出的值误差会相对大，此时步骤2906至步骤2908则通过实时数据对初始库位配额进行修正。为了保 证为每个租赁公司的每个中心仓配置的库位数能够满足其接下来N天的车辆交付，通过近N天的实际交付车辆数进行数据修正。具体逻辑如下：在初始配额下的交付峰值，即y＝X*N/A，其中X为初始库位配额，y为预测的交付峰值，A为车位流速，优选以每个车位2天1台车的流速计算；然后再与近N天租赁公司该中心仓真实的交付峰值进行比较，如果预测的交付峰值-租赁公司该中心仓真实的交付峰值>差值阈值，则不做任何修正，否则进行配额修正，修正的计算公式为：(租赁公司该中心仓真实的交付峰值+差值阈值)/(N/A)，优选向上取整得到修正后的库位配额数。公式中之所以除以A是和业务背景相关，一辆车出库和入库所需要一定的时间，通常情况下一个车在车位流速A天内周转一辆车，该模块的目的是保证基于模型预测的分配的库位数够符合每个租赁公司真实的车辆周转率。
本实施例通过预测模型预测租赁公司的总租车数量和总还车数量，然后基于每个中心仓的历史占比，将总租车数量和总还车数量分配到每个中心仓中。由于采用整个租赁公司的数据进行预测，因此具有足够大的数据样本，使得整体数据准确，而通过历史占比进行分配也能准确反映每个中心仓的情况。最后，通过对库位配额的修正，避免预测出现较大误差，同时，引入车位流速进行模型预测，符合每个租赁公司真实的车辆周转率。
图30是根据本申请一些实施例提供的一种车辆调配方法实现的示例性系统原理图。
如前所述，获取模块501可以用于负责获取各方来源数据，并做数据清洗、特征处理和结构化存储。在一些实施例中，获取模块501可以包括仓位获取装置3011、租赁公司数据获取装置3012、历史单量获取装置3013、天气节假日数据采集装置3014。
如前所述，第一确定模块502可以用于通过预测模型进行租车需求预测和还车数量预测，为后续模块提供决策数据支撑。在一些实施例中，第一确定模块50可以包括租车需求预测模块3021、以及还车数量预测模块3022。
如前所述，第二确定模块503基于预测的数据以及业务场景，做自动化的实时决策。在一些实施例中，第二确定模块503可以包括库位配额决策模块3031、以及库位配额修正模块3032。
如图30所示，第一层是数据获取层3010，其主要负责获取各方来源数据，并做数据清洗、特征处理和结构化存储，包括仓位获取装置3011、租赁公司数据获取装置3012、历史单量获取装置3013、天气节假日数据采集装置3014；第二层是基础模型预测层3020，其是通过预测模型进行租车需求预测和还车数量预测，为后续模块提供决策数据支撑，包括租车需求预测模块3021、以及还车数量预测模块3022；第三层是业务决策层3030，其基于预测的数据以及业务场景，做自动化的实时决策，包括库位配额决策模块3031、以及库位配额修正模块3032。其中：
租车需求预测模块3021：通过输入历史租车单量、CP公司规模数据、节假日信息、天气、CP公司信用分、车型丰富度、CP品牌知名度等维度特征到预测模型中，做租车单量的回归，输出每个CP公司在未来X个小时内出租车辆的预测数据。(X小时可灵活变，为一个参数)
还车数量预测模块3022：同上，通过历史老用户租期到后还车的单量数据，还CP公司的一些特征及上下文信息，将其数据处理后输入到预测模型中进行还车数量的预测，通过模型预测输出每个CP公司在未来X个小时内还车数量的预测数据。
库位配额决策模块3031：基于上述预测的租车需求数据和还车数量，可以知道每个CP公司要出租出去的车和要收回来的车，那么再将该数据按历史占比分配到每个中心仓，即可知道每个CP公司对应到每个中心仓的租车数量和还车数量，那么再结合每个CP公司在中心仓现有停放的车辆数，即可计算出每家CP公司的缺口车辆数。此时，可计算出还需为每个CP公司新增或减少多少个库位。
库位配额修正模块3032：当业务规模小时，若历史数据量级较小时，模 型预测出的值误差会相对大，此时可以通过实时数据来进行矫正。为了保证为每个CP配置的库位数能够满足其接下来N天的车辆交付，我们通过近N天的实际交付车辆数进行数据修正。具体逻辑如下：以每个车位2天1台车的流速计算在初始配额下的交付峰值，即y＝X*N/2，其中X为“库位配额决策”模块计算出的初始库位数，y为预测的交付峰值；然后再与近两N天每个CP真实的交付峰值进行比较，如果预测的交付峰值-CP真实的交付峰值>threshold，则不做任何修正，否则进行配额修正，修正的计算公式为：(CP真实的交付峰值+threshold)/(N/2)向上取整得到修正后的库位配额数。公式中之所以除以2是和业务背景相关，一辆车出库和入库所需要一定的时间，通常情况下一个车位在2天内周转一辆车，该模块的目的是保证基于模型预测的分配的库位数够每个CP公司真实的车辆周转率。
通过该技术方案实现了租车业务中心仓的库位智能化管理，能够每天动态的为每个CP公司分配库位，并且支持中心仓数量和CP数量的增加以及每个中心仓的总库位的变动，进而大规模的扩展到其他城市，只要提供仓库的一些基础信息和订单数据即可快速拓展到其他城市。其次，该方法极大的节省了运营人力，并且提高了车位利用率和车辆周转率，充分满足CP的需求，不会发生因为库位分配不公或者不够导致CP没能及时满足用户的订车需求，使得平台与CP的合作更加高效。
本申请实施例提供了一种车辆调配设备，如图31所示，该设备3100包括：
至少一个处理器3101；以及，
与所述至少一个处理器3101通信连接的存储器3102；其中，
所述存储器3102存储有可被所述一个处理器执行的指令，所述指令被所述至少一个处理器执行，以使所述至少一个处理器能够：
获取租赁公司在每个中心仓的缺口车辆数作为每个中心仓的初始库位配额；
对于每个中心仓，获取该中心仓历史时间段内的真实交付峰值，根据所述初始库位配额计算相同历史时间段内该中心仓的预测交付峰值，根据所述真实交付峰值与所述预测交付峰值的差值，通过预测模型对所述初始库位配额进行修正得到该中心仓的修正后库位配额；
向所述租赁公司的每个中心仓分配对应的所述修正后库位配额的车辆。
图31中以一个处理器3101为例。
电子设备优选为租车平台的服务器。电子设备还可以包括：输入装置3103和显示装置3104。
处理器3101、存储器3102、输入装置3103及显示装置3104可以通过总线或者其他方式连接，图中以通过总线连接为例。
存储器3102作为一种非易失性计算机可读存储介质，可用于存储非易失性软件程序、非易失性计算机可执行程序以及模块，如本申请实施例中的车辆调配方法对应的程序指令/模块，例如，图28所示的方法流程。处理器3101通过运行存储在存储器3102中的非易失性软件程序、指令以及模块，从而执行各种功能应用以及数据处理，即实现上述实施例中的租车平台中心仓车位分配方法。
存储器3102可以包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需要的应用程序；存储数据区可存储根据租车平台中心仓车位分配方法的使用所创建的数据等。此外，存储器3102可以包括高速随机存取存储器，还可以包括非易失性存储器，例如至少一个磁盘存储器件、闪存器件、或其他非易失性固态存储器件。在一些实施例中，存储器3102可选包括相对于处理器3101远程设置的存储器，这些远程存储器可以通过网络连接至执行租车平台中心仓车位分配方法的装置。上述网络的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
输入装置3103可接收输入的用户点击，以及产生与租车平台中心仓车位分配方法的用户设置以及功能控制有关的信号输入。显示装置3104可包括显示屏等显示设备。
在所述一个或者多个模块存储在所述存储器3102中，当被所述一个或者多个处理器3101运行时，执行上述任意方法实施例中的租车平台中心仓车位分配方法。
本实施例实现了租车业务中心仓的库位智能化管理，能够动态的为每个租赁公司分配库位，并且支持中心仓数量和租赁数量的增加以及每个中心仓的总库位的变动，进而大规模的扩展到其他城市，只要提供仓库的一些基础信息和订单数据即可快速拓展到其他城市。其次，该方法极大的节省了运营人力，并且提高了车位利用率和车辆周转率，充分满足租赁的需求，不会发生因为库位分配不公或者不够导致租赁没能及时满足用户的订车需求，使得平台与租赁的合作更加高效。最后，本发明根据真实交付峰值对库位配额进行修正，使得库位配额更为符合真实需求。
本申请一些实施例提供一种车辆调配电子设备，包括：
至少一个处理器；以及，
与所述至少一个处理器通信连接的存储器；其中，
所述存储器存储有可被所述一个处理器执行的指令，所述指令被所述至少一个处理器执行，以使所述至少一个处理器能够：
获取租赁公司的历史租车数量、租赁公司的信息维度数据、时间数据，输入预测模型，获得预测模型输出的租赁公司的总租车数量；
获取租赁公司的历史还车数量、租赁公司的信息维度数据、时间数据，输入预测模型，获得预测模型输出的租赁公司的总还车数量；
获取租赁公司每个中心仓的历史占比，确定每个中心仓的租车数量为总租车数量×该中心仓的历史占比，每个中心仓的还车数量为总还车数量×该中心仓的历史占比；
获取每个中心仓的停车数量；
基于每个中心仓的租车数量、还车数量、以及停车数量，确定每个中心仓的缺口车辆数作为每个中心仓的初始库位配额；
对于每个中心仓，获取该中心仓历史时间段内的真实交付峰值，根据所述初始库位配额计算相同历史时间段内该中心仓的预测交付峰值，所述预测交付峰值y＝X×N/A，其中，y为预测交付峰值，X为初始库位配额，N为历史时间段长度，A为车位流速；
如果预测交付峰值减去真实交付峰值的差值大于差值阈值，则以所述初始库位配额作为所述修正后库位配额；
如果预测交付峰值减去真实交付峰值的差值小于等于差值阈值，则计算所述修正后库位配额为：(真实交付峰值+差值阈值)/(N/A)；
向所述租赁公司的每个中心仓分配对应的所述修正后库位配额的车辆。
本实施例通过预测模型预测租赁公司的总租车数量和总还车数量，然后基于每个中心仓的历史占比，将总租车数量和总还车数量分配到每个中心仓中。由于采用整个租赁公司的数据进行预测，因此具有足够大的数据样本，使得整体数据准确，而通过历史占比进行分配也能准确反映每个中心仓的情况。最后，通过对库位配额的修正，避免预测出现较大误差，同时，引入车位流速进行模型预测，符合每个租赁公司真实的车辆周转率。
在一些实施例中，处理设备可以获取租赁公司的历史租车数量、天气数据、以及历史还车数量，形成组合特征序列，并将组合特征序列输入基于RNN模型的嵌入模型，以获得车流表示向量。进一步地，将车流表示向量和租赁公司的信息维度数据等特征输入预测模型，获得预测模型输出的租赁公司的总租车数量。嵌入模型和输入预测模型可以通过联合训练的方式获得。
其中，组合特征序列由若干时间点的特征组合值构成。每一时间点的特征组合值由该时间点的历史租车数量、天气数据、历史还车数据组合形成，并乘以时间权重系数。时间权重系数依时间点的远近可以有所不同，距当前较近的时间点的权重系数可以较大。在形成特征组合值时，可以对历史还车数量值进行开方处理，以缩小该值在进行租车预测时的影响。在一些实施例中，开方的指数可以在0.4和0.6之间，例如历史还车数量值的0.5次方被用于计算特征组合值。
类似方式也可以被用于预测总还车数量。
通过上述方式的组合，可以更好地反映出不同因素对于预测结果的影响，特别是这些因素之间的相互关系。例如，天气的影响是具有前后关联性的，租车还车数量的影响也是具有关联性的，基于RNN的处理，可以反映出前后时间点的关联关系，使预测结果更准确。
本申请实施例还提供一种存储介质，用于存储计算机指令，当计算机执行所述计算机指令时，用于执行前述的车辆调配方法的所有步骤。
示例性地，该车辆调配及其硬件装置可以应用于以下车辆调配场景。租车平台的后端服务器从租赁公司的服务器或其他数据源，获取租赁公司的历史周转车辆数目及多维度车辆周转影响数据；租车平台的后端服务器向中心仓的服务器发送中心仓的库位缺口数目的数据获取请求，中心仓的服务器响应于该数据获取请求，向租车平台的后端服务器输送相应数据；租车平台的服务器根据历史周转车辆数目及多维度车辆周转影响数据及库位缺口数目，通过预测模型进行数据处理，最终确定用于车辆调配的车辆调度方案，在根据车辆调度方案车辆调配过程中或完成后，实时更新数据并发送至租赁公司的服务器和中心仓的服务器，无论是租赁公司的供应商用户，还是中心仓的管理人员用户，都可以通过其终端登录租车平台的相应客户端获取查询车辆调配信息数据(如车辆调配的车型信息、调配进度信息等等)。
为了使得预测模型的预测结果的准确度和可靠性更高，可以利用目标租车点的关联特征样本数据及其关联车辆需求标签数据通过待训练模型训练得到。例如，待训练模型可以采用人工智能算法，具体地可以采用诸如决策树、随机森林、逻辑回归、支持向量机、朴素贝叶斯、K近邻算法、K均值算法、Adaboost(Boosting算法的一种)、神经网络、马尔科夫的机器学习算法。
关联特征样本数据为模型的输入，将关联特征样本数据的对应关联车辆需求标签数据(例如，由关联特征样本数据的历史关联车辆需求创建的标签数据)作为标签，模型的输出为关联车辆需求预测数据，对待训练模型进行训练。
在一些实施例中，可以在训练动作完成之后，将输出与标签进行比对，计算误差，然后模型参数进行反馈调整，直到输出与标签的误差较小且趋于稳定，即模型达到期望训练效果，将该训练后的模型作为预测模型。通过对训练模型的参数调整更新，提高了模型参数的精度，进一步提高了训练得到的预测模型的预测精确度、计算适应度和可靠性。在一些实施例中，训练过程3200可以由处理器执行。
图32是根据本申请一些实施例所示的预测模型训练过程3200示例性流程图。
在一些实施例中，预测模型可以通过以下步骤训练得到：
步骤3201，获取目标租车点的关联特征样本数据及其关联车辆需求标签数据；
步骤3202，将关联特征样本数据输入待训练模型，输出预测结果数据；
步骤3203，根据预测结果数据和关联车辆需求标签数据更新模型参数，并不断训练直至得到预测模型。
示例性地，可以选取每个预定城市的每个中心仓对应的多个维度的特征量作为关联特征样本数据，每个预定城市的每个中心仓对应的历史需求量作为关联车辆需求标签数据，通过决策树模型训练得到预测模型，预测模型训练的更多细节及相关描述请参见图19、图20，在此不再赘述。
在一些实施例中，可以通过目标租车点上传到服务器、租车平台的服务器，和/或，与目标租车点服务器、目标租车区域服务器具有数据通信连接关系的业务后台服务器，获取得到目标租车点的关联特征样本数据及其关联车辆需求标签数据。
通过采用诸如机器学习的人工智能算法或其他预测算法，来根据车辆调配方案不同应用场景需求进行相应模型训练，得到预期预测效果的预测模型，使得预测模型对多种目标租车点的关联特征数据的处理过程更高效，预测精确度和适应性更高，更能贴近用户车辆调配业务中获取的车辆需求信息。
本领域普通技术人员可以理解：实现上述各方法实施例的全部或部分步骤可以通过程序指令相关的硬件来完成。前述的程序可以存储于一计算机可读取存储介质中。该程序在执行时，执行包括上述各方法实施例的步骤；而前述的存储介质包括：ROM、RAM、磁碟或者光盘等各种可以存储程序代码的介质。
上文已对基本概念做了描述，显然，对于本领域技术人员来说，上述详细披露仅仅作为示例，而并不构成对本申请的限定。虽然此处并没有明确说明，本领域技术人员可能会对本申请进行各种修改、改进和修正。该类修改、改进和修正在本申请中被建议，所以该类修改、改进、修正仍属于本申请示范实施例的精神和范围。
同时，本申请使用了特定词语来描述本申请的实施例。如“一个实施例”、“一实施例”、和/或“一些实施例”意指与本申请至少一个实施例相关的某一特征、结构或特点。因此，应强调并注意的是，本申请中在不同位置两次或多次提及的“一实施例”或“一个实施例”或“一个替代性实施例”并不一定是指同一实施例。此外，本申请的一个或多个实施例中的某些特征、结构或特点可以进行适当的组合。
此外，除非权利要求中明确说明，本申请所述处理元素和序列的顺序、数字字母的使用、或其他名称的使用，并非用于限定本申请流程和方法的顺序。尽管上述披露中通过各种示例讨论了一些目前认为有用的发明实施例，但应当理解的是，该类细节仅起到说明的目的，附加的权利要求并不仅限于披露的实施例，相反，权利要求旨在覆盖所有符合本申请实施例实质和范围的修正和等价组合。例如，虽然以上所描述的系统组件可以通过硬件设备实现，但是也可以只通过软件的解决方案得以实现，如在现有的服务器或移动设备上安装所描述的系统。
同理，应当注意的是，为了简化本申请披露的表述，从而帮助对一个或多个发明实施例的理解，前文对本申请实施例的描述中，有时会将多种特征归并至一个实施例、附图或对其的描述中。但是，这种披露方法并不意味着本申请对象所需要的特征比权利要求中提及的特征多。实际上，实施例的特征要少于上述披露的单个实施例的全部特征。
一些实施例中使用了描述成分、属性数量的数字，应当理解的是，此类用于实施例描述的数字，在一些示例中使用了修饰词“大约”、“近似”或“大体上”来修饰。除非另外说明，“大约”、“近似”或“大体上”表明所述数字允许有±20％的变化。相应地，在一些实施例中，说明书和权利要求中使用的数值参数均为近似值，该近似值根据个别实施例所需特点可以发生改变。在一些实施例中，数值参数应考虑规定的有效数位并采用一般位数保留的方法。尽管本申请一些实施例中用于确认其范围广度的数值域和参数为近似值，在具体实施例中，此类数值的设定在可行范围内尽可能精确。
针对本申请引用的每个专利、专利申请、专利申请公开物和其他材料，如文章、书籍、说明书、出版物、文档等，特此将其全部内容并入本申请作为参考。与本申请内容不一致或产生冲突的申请历史文件除外，对本申请权利要求最广范围有限制的文件(当前或之后附加于本申请中的)也除外。需要说明的是，如果本申请附属材料中的描述、定义、和/或术语的使用与本申请所述内容有不一致或冲突的地方，以本申请的描述、定义和/或术语的使用为准。
最后，应当理解的是，本申请中所述实施例仅用以说明本申请实施例的原则。其他的变形也可能属于本申请的范围。因此，作为示例而非限制，本申请实施例的替代配置可视为与本申请的教导一致。相应地，本申请的实施例不仅限于本申请明确介绍和描述的实施例。

Claims (14)

  1. 一种车辆调配方法，其特征在于，包括通过处理器执行以下步骤：
    获取目标租车点的关联特征数据，所述关联特征数据包括所述目标租车点和目标租车区域中的至少一种的相关特征数据；所述目标租车区域为与所述目标租车点相关的区域；
    基于预测模型对所述目标租车点的关联特征数据进行处理，确定所述目标租车点的关联车辆需求，所述关联车辆需求为所述目标租车点在所述目标租车区域中对车辆的相关需求；
    至少基于所述关联车辆需求，确定所述目标租车点的车辆调度方案。
  2. 根据权利要求1所述的方法，其特征在于，所述关联特征数据包括：所述目标租车点的被租用车辆的第一历史车辆属性信息和所述目标租车区域的历史租车订单对应的第二历史车辆属性信息；所述目标租车点的关联车辆需求包括所述目标租车点的车型需求；
    所述基于预测模型对所述目标租车点的关联特征数据进行处理，确定所述目标租车点的关联车辆需求，包括：
    基于所述预测模型对所述目标租车点的被租用车辆的第一历史车辆属性信息和所述目标租车区域的历史租车订单对应的第二历史车辆属性信息进行处理，确定所述目标租车点的车型需求。
  3. 根据权利要求2所述的方法，其特征在于，所述至少基于所述关联车辆需求，确定所述目标租车点的车辆调度方案，包括：
    基于待调度车辆的第三车辆属性信息和所述目标租车点的车型需求，从所述待调度车辆中确定所述车辆调度方案的调度车辆。
  4. 根据权利要求3所述的方法，其特征在于，所述第一历史车辆属性信息包括所述目标租车点的各个被租用车辆对应的历史车型信息，所述第二历史车 辆属性信息包括所述目标租车区域的各所述历史租车订单包括的租用车辆的车型信息，
    所述基于所述预测模型对所述目标租车点的被租用车辆的第一历史车辆属性信息和所述目标租车区域的历史租车订单对应的第二历史车辆属性信息进行处理，确定所述目标租车点的车型需求，包括：
    基于预测模型对所述目标租车点各个被租用车辆对应的历史车型信息、各所述历史租车订单包括的租用车辆的车型信息进行处理，确定所述目标租车点的目标车型及其对应的目标车型数目；
    所述基于待调度车辆的第三车辆属性信息和所述目标租车点的车型需求，从所述待调度车辆中确定所述车辆调度方案的调度车辆，包括：
    基于所述待调度车辆的第三车辆属性信息和所述目标租车点的目标车型及其对应的目标车型数目，选定与所述目标租车点的目标车型匹配、所述目标车型数目的调度车辆。
  5. 根据权利要求1所述的方法，其特征在于，所述关联特征数据包括：所述目标租车点的环境信息和所述目标租车区域的选品相关特征数据；所述目标租车点的关联车辆需求包括所述目标租车点的车型需求；
    所述基于预测模型对所述目标租车点的关联特征数据进行处理，确定所述目标租车点的关联车辆需求，包括：
    基于所述预测模型对所述目标租车点的环境信息和所述目标租车区域的选品相关特征数据进行处理，确定所述目标租车点的车型需求。
  6. 根据权利要求5所述的方法，其特征在于，所述至少基于所述关联车辆需求，确定所述目标租车点的车辆调度方案，包括：
    基于待调度车辆的第三车辆属性信息和所述目标租车点的车型需求，从所述待调度车辆中确定所述车辆调度方案的调度车辆。
  7. 根据权利要求6所述的方法，其特征在于，所述目标租车区域的选品相关特征数据包括所述目标租车区域多个维度的历史选品需求相关数据，
    基于所述预测模型对所述目标租车点的环境信息和所述目标租车区域的选品相关特征数据进行处理，确定所述目标租车点的车型需求，包括：
    基于所述预测模型对所述目标租车点的环境信息和所述目标租车区域多个维度的历史选品需求相关数据进行处理，确定所述目标租车点的目标车型及其对应的目标车型数目；
    所述基于待调度车辆的第三车辆属性信息和所述目标租车点的车型需求，从所述待调度车辆中确定所述车辆调度方案的调度车辆，包括：
    基于所述待调度车辆的第三车辆属性信息和所述目标租车点的目标车型及其对应的目标车型数目，选定与所述目标租车点的目标车型匹配、所述目标车型数目的调度车辆。
  8. 根据权利要求1所述的方法，其特征在于，所述关联特征数据包括：所述目标租车点的历史周转车辆信息及相关信息、所述目标租车区域的库位信息；所述目标租车点的关联车辆需求包括所述目标租车点的车辆数需求；
    所述基于预测模型对所述目标租车点的关联特征数据进行处理，确定所述目标租车点的关联车辆需求，包括：
    基于预测模型对所述目标租车点的历史周转车辆信息及相关信息、所述目标租车区域的库位信息进行处理，确定所述目标租车点的车辆数需求。
  9. 根据权利要求8所述的方法，其特征在于，所述至少基于所述关联车辆需求，确定所述目标租车点的车辆调度方案，包括：
    基于待调度车辆的第三车辆属性信息和所述目标租车点的车辆数需求，从所述待调度车辆中确定所述车辆调度方案的调度车辆。
  10. 根据权利要求9所述的方法，其特征在于，所述历史周转车辆信息及相关信息包括历史周转车辆数目及多维度车辆周转影响数据，所述库位信息包括库位缺口数目，
    所述基于预测模型对所述目标租车点的历史周转车辆信息及相关信息、所述目标租车区域的库位信息进行处理，确定所述目标租车点的车辆数需求，包括：
    基于所述预测模型对所述目标租车点的历史周转车辆数目及多维度车辆周转影响数据、所述目标租车区域的库位缺口数目进行处理，确定所述目标租车点对应所述目标租车区域的车辆数配额需求；
    所述基于待调度车辆的第三车辆属性信息和所述目标租车点的车辆数需求，从所述待调度车辆中确定所述车辆调度方案的调度车辆，包括：
    基于待调度车辆的第三车辆属性信息和所述目标租车点的车辆数配额需求，从所述待调度车辆中确定所述车辆调度方案的调度车辆。
  11. 根据权利要求1所述的方法，其特征在于，所述预测模型通过以下过程训练得到：
    获取所述目标租车点的关联特征样本数据及其关联车辆需求标签数据；
    将所述关联特征样本数据输入待训练模型，输出预测结果数据；
    根据所述预测结果数据和所述关联车辆需求标签数据更新模型参数，并不断训练直至得到所述预测模型。
  12. 一种车辆调配装置，其特征在于，包括处理器，所述处理器包括：
    获取模块，用于：获取目标租车点的关联特征数据，所述关联特征数据包括所述目标租车点和目标租车区域中的至少一种的相关特征数据；所述目标租车区域为与所述目标租车点相关的区域；
    第一确定模块，用于：基于预测模型对所述目标租车点的关联特征数据进行 处理，确定所述目标租车点的关联车辆需求，所述关联车辆需求为所述目标租车点在所述目标租车区域中对车辆的相关需求；
    第二确定模块，用于：至少基于所述关联车辆需求，确定所述目标租车点的车辆调度方案。
  13. 一种车辆调配设备，其特征在于，包括：
    存储器；
    处理器；以及
    计算机程序；
    其中，所述计算机程序存储在所述存储器中，并被配置为由所述处理器执行以实现如权利要求1-11中任一所述的方法。
  14. 一种计算机可读存储介质，其上存储有计算机程序，其特征在于，所述计算机程序被处理器执行时实现如权利要求1-11任一项所述的方法。
PCT/CN2020/139589 2019-12-26 2020-12-25 一种车辆调配方法、装置、设备及计算机可读存储介质 WO2021129831A1 (zh)

Applications Claiming Priority (6)

Application Number Priority Date Filing Date Title
CN201911369388.1 2019-12-26
CN201911369388.1A CN111815012A (zh) 2019-12-26 2019-12-26 租车平台中心仓车位分配方法、电子设备及存储介质
CN202010006858.4A CN111832872A (zh) 2020-01-03 2020-01-03 一种车辆调度方法、装置、电子设备和存储介质
CN202010006858.4 2020-01-03
CN202010171414.6 2020-03-12
CN202010171414.6A CN111383055B (zh) 2020-03-12 2020-03-12 租车中心仓选品方法、装置、设备及计算机可读存储介质

Publications (1)

Publication Number Publication Date
WO2021129831A1 true WO2021129831A1 (zh) 2021-07-01

Family

ID=76573698

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/139589 WO2021129831A1 (zh) 2019-12-26 2020-12-25 一种车辆调配方法、装置、设备及计算机可读存储介质

Country Status (1)

Country Link
WO (1) WO2021129831A1 (zh)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170075358A1 (en) * 2014-05-06 2017-03-16 Huawei Technologies Co., Ltd. Self-driving car scheduling method, car scheduling server, and self-driving car
CN108510730A (zh) * 2018-04-24 2018-09-07 徐漫洋 共享汽车网点间车辆调度方法及系统
CN108596484A (zh) * 2018-04-24 2018-09-28 徐漫洋 一种基于网点属性分析的共享汽车调度方法和系统
CN110400078A (zh) * 2019-07-25 2019-11-01 百度在线网络技术（北京）有限公司 无人驾驶车辆调度方法及装置
CN111383055A (zh) * 2020-03-12 2020-07-07 北京嘀嘀无限科技发展有限公司 租车中心仓选品方法、装置、设备及计算机可读存储介质
CN111815012A (zh) * 2019-12-26 2020-10-23 北京嘀嘀无限科技发展有限公司 租车平台中心仓车位分配方法、电子设备及存储介质
CN111832872A (zh) * 2020-01-03 2020-10-27 北京嘀嘀无限科技发展有限公司 一种车辆调度方法、装置、电子设备和存储介质

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170075358A1 (en) * 2014-05-06 2017-03-16 Huawei Technologies Co., Ltd. Self-driving car scheduling method, car scheduling server, and self-driving car
CN108510730A (zh) * 2018-04-24 2018-09-07 徐漫洋 共享汽车网点间车辆调度方法及系统
CN108596484A (zh) * 2018-04-24 2018-09-28 徐漫洋 一种基于网点属性分析的共享汽车调度方法和系统
CN110400078A (zh) * 2019-07-25 2019-11-01 百度在线网络技术（北京）有限公司 无人驾驶车辆调度方法及装置
CN111815012A (zh) * 2019-12-26 2020-10-23 北京嘀嘀无限科技发展有限公司 租车平台中心仓车位分配方法、电子设备及存储介质
CN111832872A (zh) * 2020-01-03 2020-10-27 北京嘀嘀无限科技发展有限公司 一种车辆调度方法、装置、电子设备和存储介质
CN111383055A (zh) * 2020-03-12 2020-07-07 北京嘀嘀无限科技发展有限公司 租车中心仓选品方法、装置、设备及计算机可读存储介质

Similar Documents

Publication Publication Date Title
US10639995B2 (en) Methods, circuits, devices, systems and associated computer executable code for driver decision support
CN110782064B (zh) 一种集车辆调度优化与任务分配于一体的可视化方法及系统
CN104598979B (zh) 基于时间和位置的递送最优化
CN107437144A (zh) 一种订单调度方法、系统、计算机设备和存储介质
KR20180011053A (ko) 운송 능력 스케줄링을 위한 방법들 및 시스템들
CN108874801B (zh) 一种配送站选址的方法和装置
US10789558B2 (en) Non-linear systems and methods for destination selection
CN111915254A (zh) 一种适用于汽车售后配件的库存优化控制方法和系统
CN110444008B (zh) 一种车辆调度方法及装置
CN111695842B (zh) 配送方案确定方法、装置、电子设备及计算机存储介质
CN111861296A (zh) 一种揽件任务分配方法、装置、揽件系统、设备及介质
CN114298559A (zh) 换电站的换电方法、换电管理平台及存储介质
US20140214715A1 (en) Scheduling system and method for distribution of perishable loads of pre-mixed concrete to multiple sites
CN113538067A (zh) 一种基于机器学习的城际网约车需求预测方法及系统
CN111383055B (zh) 租车中心仓选品方法、装置、设备及计算机可读存储介质
CN113642761A (zh) Robotaxi自动驾驶共享网约车资源分配方法
WO2021129831A1 (zh) 一种车辆调配方法、装置、设备及计算机可读存储介质
CN111429237A (zh) 订单价格的确定方法、装置、服务器及存储介质
CN112906980A (zh) 订单处理方法、装置、系统及可读存储介质
WO2021233958A1 (en) Quote prediction
CN110414875B (zh) 产能数据处理方法、装置、电子设备及计算机可读介质
CN111047354A (zh) 一种分时定价的实现方法、装置、电子设备和存储介质
CN111612198A (zh) 预测拼单成功率的方法、装置和电子设备
CN115032957B (zh) 生产调度方法及装置、存储介质、电子设备
US20240062146A1 (en) Hybrid system for multi-order batching with real-time and item awareness

Legal Events

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

Ref document number: 20904676

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20904676

Country of ref document: EP

Kind code of ref document: A1