Scheduling method and device for electric bicycles

Abstract

The invention provides an electric bicycle scheduling method and device, and the method comprises the steps: predicting and obtaining an initial scheduling task based on a scheduling task prediction model application multi-dimensional operation data, and weighting the execution priority of the initial scheduling task based on the prediction order turnover data of an electric bicycle station, and obtaining a revenue calibration scheduling task; on the basis of the scheduling parameters of the historical scheduling tasks of the scheduling types, weighting the execution priorities of the income calibration scheduling tasks of the scheduling types to obtain management calibration scheduling tasks; and scheduling the electric bicycles based on the management calibration scheduling task. According to the method provided by the invention, electric bicycle scheduling for comprehensively improving riding income and scheduling efficiency is realized, personal experience interference is avoided, secondary planning is carried out manually, and the rationality of a scheduling task is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electric motorcycles, and in particular to a dispatching method and device for electric motorcycles. Background Art

[0002] The traditional dispatching method relies on the driver's independent decision and dispatches to popular stations based on experience. There are unreasonable aspects in the dispatched stations and the number of vehicles. In recent years, most service providers have been moving towards intelligence, with scheduling tasks decided by systems and algorithms. Currently, most of them use big data and AI algorithms to input user riding data and riding point data in the area, infer user vehicle demand at points in the area, and thus guide service providers to dispatch, better meet market riding needs and increase revenue.

[0003] However, most of the current industry solutions are only aimed at tasks aimed at increasing revenue, and lack intelligent scheduling capabilities for other tasks such as urban parking regulation, strict regional scheduling, asset protection, etc. Summary of the invention

[0004] The present invention provides a dispatching method and device for electric motorcycles, which are used to solve the defect that the dispatching scheme in the prior art is difficult to achieve the dispatching task for the purpose of improving revenue and management efficiency.

[0005] The present invention provides a dispatching method for electric motorcycles, comprising: Obtain multi-dimensional operational data of motorcycle stations; Applying the multi-dimensional operation data based on a scheduling task prediction model to predict the initial scheduling task of the motorcycle station, wherein the scheduling task prediction model is constructed based on a machine learning algorithm; Based on the predicted order turnover data of the motorcycle station, weighting the execution priority of the initial scheduling task to obtain a revenue calibration scheduling task; Based on the scheduling parameters of the historical scheduling tasks of each scheduling type, weighting the execution priority of the revenue calibration scheduling tasks of each scheduling type, to obtain the management calibration scheduling task; The motorcycles are dispatched based on the management calibration scheduling task.

[0006] According to a motorcycle dispatching method provided by the present invention, the dispatching parameters of the historical dispatching tasks of each dispatching type are weighted to obtain the management calibration dispatching tasks, including: Based on the scheduling parameters of the historical scheduling tasks of each scheduling type, a single-vehicle scheduling route between a popular vehicle-deficient station and an unpopular station is obtained; Fitting the dispatching start and end points of the bicycle dispatching routes of each dispatching type to obtain the fitted dispatching routes of each dispatching type; Based on the scheduling efficiency of the fitting scheduling route of each scheduling type, the execution priority of the revenue calibration scheduling task of each scheduling type is weighted to obtain the management calibration scheduling task.

[0007] According to a method for dispatching electric motorcycles provided by the present invention, the dispatching type includes at least one of a revenue task, an inspection task, an operation task, and a safety management task.

[0008] According to a motorcycle dispatching method provided by the present invention, the execution priority of the initial dispatching task is weighted based on the predicted order turnover data of the motorcycle station to obtain the revenue calibration dispatching task, including: Determine high-yield stations based on the predicted order turnover data of the motorcycle stations; The execution priorities of the initial scheduling tasks corresponding to the high-yield sites are weighted to obtain the benefit calibration scheduling tasks.

[0009] According to a method for dispatching electric motorcycles provided by the present invention, the prediction parameters of the dispatching task prediction model include at least one of user riding demand at a station, a dispatching trigger condition, a task generation time at a station, a station priority, and a real-time supply and demand situation at a station; The prediction parameters of the scheduling task prediction model are determined based on the scheduling type.

[0010] According to a method for dispatching an electric motorcycle provided by the present invention, after dispatching the electric motorcycle based on the management calibration dispatch task, the method comprises: Obtaining dispatch effect data of the motorcycle station, wherein the dispatch effect data includes the first order generation time of the motorcycle station and / or the predicted order turnover data; Based on the scheduling effect data, the prediction parameters are optimized.

[0011] The present invention also provides a dispatching device for an electric motorcycle, comprising: An acquisition unit, for acquiring multi-dimensional operation data of an electric motorcycle station; A prediction unit, applying the multi-dimensional operation data based on a scheduling task prediction model to predict an initial scheduling task for the motorcycle station, wherein the scheduling task prediction model is constructed based on a machine learning algorithm; A revenue calibration unit, based on the predicted order turnover data of the motorcycle station, weights the execution priority of the initial scheduling task to obtain a revenue calibration scheduling task; A management calibration unit, based on the scheduling parameters of the historical scheduling tasks of each scheduling type, weights the execution priority of the revenue calibration scheduling tasks of each scheduling type to obtain a management calibration scheduling task; The task dispatching unit dispatches the motorcycles based on the management calibration scheduling tasks.

[0012] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the dispatching method for an electric motorcycle as described above is implemented.

[0013] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described methods for dispatching electric motorcycles.

[0014] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the dispatching method of an electric motorcycle as described in any one of the above-mentioned methods is implemented.

[0015] The motorcycle dispatching method and device provided by the present invention apply multi-dimensional operation data through a dispatching task prediction model to predict the initial dispatching task of the motorcycle station, and weight the execution priority of the initial dispatching task based on the predicted order turnover data of the motorcycle station to obtain the revenue calibration dispatching task; based on the dispatching parameters of the historical dispatching tasks of each dispatching type, weight the execution priority of the revenue calibration dispatching task of each dispatching type to obtain the management calibration dispatching task; dispatch the motorcycles based on the management calibration dispatching task, thereby realizing the dispatching of motorcycles to comprehensively improve the riding revenue and dispatching efficiency, avoiding the interference of personal experience and manual secondary planning, and improving the rationality of the dispatching tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 It is one of the flow charts of the dispatching method of electric motorcycles provided by the present invention; Figure 2 This is the second flow chart of the dispatching method for electric motorcycles provided by the present invention; Figure 3 It is a structural schematic diagram of the dispatching device of an electric motorcycle provided by the present invention; Figure 4It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] It should be noted that the operation scenarios of electric vehicles are complex. In addition to dispatching the car demand of C-end users for the purpose of increasing revenue, there are also dispatching for urban standardized parking, dispatching in strictly controlled areas, and dispatching for asset protection. However, most of the current industry solutions are only for tasks aimed at increasing revenue, and lack intelligent dispatching capabilities for other scenarios.

[0020] This is because the existing intelligent scheduling does not integrate the scheduling tasks well. Since most intelligent scheduling only generates revenue-increasing tasks, it is separated from other types of tasks. Frontline personnel need to think twice to integrate and plan different tasks and routes. Task execution is inefficient and requires more time and manual management to ensure the reasonable execution of tasks. In other words, the current intelligent scheduling solution is difficult to achieve a scheduling method for the purpose of increasing revenue and management efficiency.

[0021] In response to the above problems, the present invention provides a dispatching method for electric motorcycles to achieve intelligent dispatching for the purpose of improving revenue and management efficiency. Figure 1 It is one of the flow charts of the dispatching method of motorcycles provided by the present invention, such as Figure 1 As shown, the method includes: Step 110, obtaining multi-dimensional operation data of the motorcycle station; Here, the motorcycle station refers to all motorcycle stations within a certain area, and the number of motorcycle stations includes at least 1. In addition, the multi-dimensional operation data here can be used to reflect the data information of the user riding behavior dimension, environmental dimension, and station dimension, and may include the current date type, weather information, temperature information corresponding to different vehicle use periods, regional information, motorcycle station attribute information, user historical riding order information, the number of order turnovers on the day, the number of continuous order turnovers, and historical scheduling data.

[0022] It should be noted that the current date type can be divided into weekdays, weekends, and holidays; weather information can include current weather information, historical weather information, and future predicted weather information; users' historical riding order information, the number of order turnovers on the day, and the number of continuous order turnovers can be used to reflect the user riding behavior at the motorcycle station.

[0023] Specifically, the current date type, weather information, temperature information corresponding to different vehicle use periods, regional information, motorcycle station attribute information, user historical riding order information, daily order turnover times, order continuous turnover times, and historical dispatch data of motorcycle stations in a certain area can be obtained. Then, the initial data of each motorcycle station can be processed, and the processed data can be used as the multi-dimensional operation data of each motorcycle station, as the basic data for subsequent dispatch task prediction analysis of each motorcycle station.

[0024] In one embodiment, the initial multi-dimensional operation data of each motorcycle station can be obtained by obtaining the current date type, weather information, temperature information corresponding to different vehicle use periods, regional information, motorcycle station attribute information, user historical riding order information, the number of order turnovers on the day, the number of continuous order turnovers, and historical scheduling data of the motorcycle stations in a certain area. Then, the initial multi-dimensional operation data of each motorcycle station can be cleaned to remove invalid data, such as zero order data caused by abnormal weather and a large amount of order data caused by social activities. Then, after data cleaning, the cleaned data of each motorcycle station is integrated to form grid data. In detail, the cleaned data can be automatically judged by an algorithm. For example, if a user starts using the car at multiple stations, the distance between the multiple stations is within a certain distance, and a certain frequency is met, then the multiple stations can be integrated into one station, and the cleaned data of the multiple stations are integrated into the grid data of one station. Here, there is an upper limit to the number of neighboring station groups in the grid data.

[0025] It should be noted that in the process of obtaining multi-dimensional operation data of electric motorcycles, the logic of adjacent stations is added, and the vehicle data of adjacent stations and the vehicle data of nearby vehicles that are not at the stations are placed in the scheduling task prediction model, avoiding repeated scheduling of starting and ending points and improving the scheduling capacity of non-station vehicles.

[0026] Step 120, applying the multi-dimensional operation data based on a scheduling task prediction model to predict the initial scheduling task of the motorcycle station, wherein the scheduling task prediction model is constructed based on a machine learning algorithm; Here, the scheduling task prediction model is constructed based on a machine learning algorithm and can be pre-trained for the purpose of increasing revenue.

[0027] In addition, the prediction parameters in the scheduling task prediction model may be divided based on the scheduling type, and the specific values ​​of the prediction parameters corresponding to different scheduling types may be different.

[0028] Specifically, first, the current general scheduling task prediction model can be obtained, and the multi-dimensional operation data can be input into the scheduling task prediction model. The scheduling task prediction model can be used to output the number of in-and-out orders per day and per hour for each electric vehicle station. Then, the number of in-and-out orders per day and per hour and the real-time number of vehicles at the motorcycle station can be predicted to calculate the number of vehicles transferred in or out of each motorcycle station, that is, the initial scheduling task of each motorcycle station can be obtained. In addition, the motorcycle stations that need to transfer in a large number of vehicles can be regarded as short-car stations, and the motorcycle stations that need to transfer out a large number of vehicles can be regarded as accumulation stations.

[0029] Step 130, based on the predicted order turnover data of the motorcycle station, weighting the execution priority of the initial scheduling task to obtain a revenue calibration scheduling task; Here, the predicted order turnover data refers to the number of order turnovers generated by the motorcycles affected by the initial scheduling task in the future time period after the initial scheduling task is performed. Therefore, the predicted order turnover data can be used to reflect the impact of the initial scheduling task on the number of orders at the site, that is, to reflect the impact of the initial scheduling task on the revenue.

[0030] Specifically, after obtaining the initial dispatching task of the motorcycle station, the predicted order turnover data of the motorcycle station after executing the initial dispatching task can be predicted through deep prediction. Then, the execution priority of the initial dispatching task of each motorcycle station can be weighted according to the amount of order turnover data reflected by the predicted order turnover data, and the weighted initial dispatching task can be used as the revenue calibration dispatching task.

[0031] It can be understood that the more order turnover data reflected by the predicted order turnover data, the more revenue the initial scheduling task brings. The execution priority of the initial scheduling task of the motorcycle station can be weighted so that the initial scheduling task of the motorcycle station is executed first, that is, a revenue-calibrated scheduling task adjusted based on the revenue is obtained.

[0032] Step 140, weighting the execution priority of the initial scheduling tasks of each scheduling type based on the scheduling parameters of the historical scheduling tasks of each scheduling type to obtain a management calibration scheduling task; Here, the scheduling type may include at least one of a revenue task, an inspection task, an operation task, and a safety management task. In addition, the scheduling parameters include parameters related to the scheduling task, such as the site scheduling direction, the distance between sites, and the scheduling route.

[0033] Specifically, the management calibration scheduling task can be obtained by analyzing the scheduling parameters of the historical scheduling tasks of each scheduling type and weighting the execution priority of the initial scheduling tasks of each scheduling type. It can be understood that the scheduling efficiency of the scheduling tasks of each scheduling type can be obtained by analyzing the scheduling parameters of the historical scheduling tasks of each scheduling type. Therefore, by analyzing the scheduling parameters of the historical scheduling tasks of each scheduling type, the initial scheduling tasks of the scheduling type with lower cost corresponding to the scheduling distance and scheduling route can be weighted, and the revenue calibration scheduling tasks of the scheduling type with low cost can be executed first, and then the scheduling tasks with weighted execution priority of each revenue calibration scheduling task can be used as the management calibration scheduling task.

[0034] It can be understood that the obtained management calibration scheduling task can realize both high-yield task scheduling and high-efficiency task scheduling, thereby improving the comprehensive level of the scheduling tasks to be executed.

[0035] Step 150, dispatching the motorcycles based on the management calibration scheduling task.

[0036] Specifically, the corresponding operation and maintenance personnel can be arranged according to the management calibration scheduling tasks to perform the scheduling of each motorcycle station. For example, the accumulated vehicles at unpopular stations can be dispatched to popular stations according to the scheduling routes.

[0037] The method provided by the embodiment of the present invention applies multi-dimensional operation data through a scheduling task prediction model to predict the initial scheduling task of a motorcycle station, and weights the execution priority of the initial scheduling task based on the predicted order turnover data of the motorcycle station to obtain a revenue calibration scheduling task; based on the scheduling parameters of the historical scheduling tasks of each scheduling type, weights the execution priority of the revenue calibration scheduling task of each scheduling type to obtain a management calibration scheduling task; scheduling of motorcycles is performed based on the management calibration scheduling task, thereby realizing the scheduling of motorcycles to comprehensively improve riding revenue and scheduling efficiency, avoiding interference from personal experience and manual secondary planning, and improving the rationality of scheduling tasks.

[0038] Based on any of the above embodiments, step 140 includes: Based on the scheduling parameters of the historical scheduling tasks of each scheduling type, a single-vehicle scheduling route between a popular vehicle-deficient station and an unpopular station is obtained; Fitting the dispatching start and end points of the bicycle dispatching routes of each dispatching type to obtain the fitted dispatching routes of each dispatching type; Based on the scheduling efficiency of the fitting scheduling route of each scheduling type, the execution priority of the revenue calibration scheduling task of each scheduling type is weighted to obtain the management calibration scheduling task.

[0039] Specifically, first, the single-vehicle dispatching route between the popular vehicle-shortage stations and the unpopular stations can be calculated from the dispatching parameters of the historical dispatching tasks of each dispatching type. Then, by fitting the dispatching start and end points of the single-vehicle dispatching route of each dispatching type in the single-vehicle dispatching route between the popular vehicle-shortage stations and the unpopular stations, the fitted dispatching route of each dispatching type can be obtained. Finally, the dispatching efficiency of the fitted dispatching route of each dispatching type can be determined by the time cost, manpower cost, and material cost of the fitted dispatching route of each dispatching type. Finally, the execution priority of the revenue calibration dispatching task of the corresponding dispatching type can be weighted by the dispatching efficiency of each dispatching type to obtain the management calibration dispatching task.

[0040] It should be noted that when scheduling operation and maintenance tasks, scheduling efficiency is used as an important reference factor for scheduling task generation. Calibration is performed through task distance and direction factors to ensure that tasks can be completed at low cost and scheduling efficiency is improved.

[0041] Based on any of the above embodiments, the scheduling type includes at least one of a revenue task, an inspection task, an operation task, and a safety management task.

[0042] Based on any of the above embodiments, step 130 includes: Determine high-yield stations based on the predicted order turnover data of the motorcycle stations; The execution priorities of the initial scheduling tasks corresponding to the high-yield sites are weighted to obtain the benefit calibration scheduling tasks.

[0043] Specifically, first, the predicted order turnover data of each motorcycle station can be obtained through prediction. Then, the stations with higher order turnover data can be regarded as high-profit stations. Then, the execution priority of the initial scheduling tasks corresponding to the high-profit stations is weighted to obtain the profit calibration scheduling tasks with adjusted execution priority, that is, the execution priority of the initial scheduling tasks corresponding to the high-profit stations is increased, so that the initial scheduling tasks corresponding to the high-profit stations can be executed first.

[0044] Based on any of the above embodiments, the prediction parameters of the scheduling task prediction model include at least one of user riding demand of the site, scheduling trigger conditions, task generation time of the site, site priority, and real-time supply and demand conditions of the site; The prediction parameters of the scheduling task prediction model are determined based on the scheduling type.

[0045] It should be noted that the prediction parameters of the scheduling task prediction model here are determined based on the scheduling type, and the specific values ​​of the prediction parameters corresponding to each scheduling type may be different. Therefore, when the scheduling task model is applied with multi-dimensional operation data to generate the scheduling task of the corresponding scheduling type, the scheduling task is obtained by predicting the prediction parameters corresponding to each scheduling type.

[0046] In the actual application of the scheduling task prediction model, different scheduling types correspond to different task scenarios. By inputting the multi-dimensional operation data of each motorcycle station into the scheduling task prediction model, the scheduling task prediction model can simultaneously output the scheduling tasks corresponding to each scheduling type, without the need for front-line personnel to conduct secondary thinking and planning tasks.

[0047] Based on any of the above embodiments, after step 150, the following steps are included: Obtaining dispatch effect data of the motorcycle station, wherein the dispatch effect data includes the first order generation time of the motorcycle station and / or the predicted order turnover data; Based on the scheduling effect data, the prediction parameters are optimized.

[0048] Specifically, after the motorcycles are dispatched according to the management calibration dispatch task, the dispatch effect data of the dispatched motorcycle stations can be obtained. For example, the first order generation time and / or predicted order turnover data of the motorcycle stations can be obtained.

[0049] Among them, the first order generation time refers to the time when the first riding order of the motorcycle is generated due to the execution of the scheduling task. It can be understood that the closer the first order generation time is to the time of executing the scheduling task, the better the effect of the scheduling task is, and the greater the help to the riding income of the motorcycle. Similarly, the larger the order turnover reflected by the predicted order turnover data, the better the effect of the scheduling task is, and the greater the help to the riding income of the motorcycle.

[0050] Finally, the prediction parameters of the scheduling task prediction model can be optimized through the scheduling effect data. In particular, the prediction parameters of a single scheduling type can be optimized to improve the prediction efficiency of the scheduling task prediction model.

[0051] Based on any of the above embodiments, Figure 2 FIG. 2 is a flow chart of the dispatching method for motorcycles provided by the present invention. Figure 2 As shown, the method includes: First, data is obtained, including date, region, weather, order, historical dispatch data, etc. Then, data processing is performed on the obtained data. In the data processing stage, the initial data obtained is first cleaned, and then the station data is calculated for the cleaned data to obtain the cleaned data of each motorcycle station. Finally, the cleaned data of the motorcycle station can be integrated through the algorithm to obtain gridded station data, and the integrated data (multi-dimensional operation data) is used as the input of the dispatch task prediction model.

[0052] The scheduling task prediction model is used to perform algorithm prediction and generate scheduling tasks. Specifically, the daily and hourly data of the station are predicted first, and then the number of trial vehicles at the station is combined to obtain the final inbound and outbound quantities of each motorcycle station, and the initial scheduling tasks of each motorcycle station are obtained.

[0053] Next, the initial dispatching tasks of each motorcycle station are calibrated. During the calibration task stage, the tasks are weighted and adjusted by obtaining data such as predicted continuous orders, trajectories, dispatching distances, directions, routes, etc. In addition, by obtaining data on changes in riding and operation and maintenance tasks at the station, the task volume is calculated and the current tasks are intervened in real time, that is, the initial dispatching tasks are intervened and adjusted in real time, and finally the task distribution of each motorcycle station is realized.

[0054] Finally, the dispatch effect data is obtained, that is, the dispatch effect data of the motorcycle station is obtained. And the prediction parameters of the dispatch task prediction model are automatically injected and iterated through dispatch.

[0055] Based on any of the above embodiments, Figure 3 Schematic diagram of the structure of the dispatching device for an electric motorcycle provided by the present invention, such as Figure 3 As shown, the device comprises: An acquisition unit 310 acquires multi-dimensional operation data of a motorcycle station; A prediction unit 320 applies the multi-dimensional operation data based on a scheduling task prediction model to predict an initial scheduling task for the motorcycle station, wherein the scheduling task prediction model is constructed based on a machine learning algorithm; The revenue calibration unit 330 weights the execution priority of the initial scheduling task based on the predicted order turnover data of the motorcycle station to obtain a revenue calibration scheduling task; The management calibration unit 340 weights the execution priority of the revenue calibration scheduling tasks of each scheduling type based on the scheduling parameters of the historical scheduling tasks of each scheduling type to obtain the management calibration scheduling tasks; The task dispatching unit 350 performs the dispatch of the motorcycle based on the management calibration dispatch task.

[0056] The device provided by the embodiment of the present invention applies multi-dimensional operating data through a scheduling task prediction model to predict the initial scheduling task of a motorcycle station, and weights the execution priority of the initial scheduling task based on the predicted order turnover data of the motorcycle station to obtain a revenue calibration scheduling task; based on the scheduling parameters of the historical scheduling tasks of each scheduling type, the execution priority of the revenue calibration scheduling task of each scheduling type is weighted to obtain a management calibration scheduling task; motorcycles are scheduled based on the management calibration scheduling task, thereby realizing motorcycle scheduling that comprehensively improves riding revenue and scheduling efficiency, avoiding interference from personal experience and manual secondary planning, and improving the rationality of scheduling tasks.

[0057] Based on any of the above embodiments, the management calibration unit is specifically used for: Based on the scheduling parameters of the historical scheduling tasks of each scheduling type, a single-vehicle scheduling route between a popular vehicle-deficient station and an unpopular station is obtained; Fitting the dispatching start and end points of the bicycle dispatching routes of each dispatching type to obtain the fitted dispatching routes of each dispatching type; Based on the scheduling efficiency of the fitting scheduling route of each scheduling type, the execution priority of the revenue calibration scheduling task of each scheduling type is weighted to obtain the management calibration scheduling task.

[0058] Based on any of the above embodiments, the scheduling type includes at least one of a revenue task, an inspection task, an operation task, and a safety management task.

[0059] Based on any of the above embodiments, the revenue calibration unit is specifically used for: Determine high-yield stations based on the predicted order turnover data of the motorcycle stations; The execution priorities of the initial scheduling tasks corresponding to the high-yield sites are weighted to obtain the benefit calibration scheduling tasks.

[0060] Based on any of the above embodiments, the prediction parameters of the scheduling task prediction model include at least one of user riding demand of the site, scheduling trigger conditions, task generation time of the site, site priority, and real-time supply and demand conditions of the site; The prediction parameters of the scheduling task prediction model are determined based on the scheduling type.

[0061] Based on any of the above embodiments, the task dispatching unit further comprises an optimization unit, wherein the optimization unit is specifically used for: Obtaining dispatch effect data of the motorcycle station, wherein the dispatch effect data includes the first order generation time of the motorcycle station and / or the predicted order turnover data; Based on the scheduling effect data, the prediction parameters are optimized.

[0062] Figure 4 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 4As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430 and a communication bus 440, wherein the processor 410, the communication interface 420 and the memory 430 communicate with each other through the communication bus 440. The processor 410 may call the logic instructions in the memory 430 to execute the dispatching method of the motorcycle, the method comprising: obtaining the multi-dimensional operation data of the motorcycle station; applying the multi-dimensional operation data based on the dispatching task prediction model to predict the initial dispatching task of the motorcycle station, the dispatching task prediction model is constructed based on the machine learning algorithm; based on the predicted order turnover data of the motorcycle station, weighting the execution priority of the initial dispatching task to obtain the revenue calibration dispatching task; based on the dispatching parameters of the historical dispatching tasks of each dispatching type, weighting the execution priority of the revenue calibration dispatching task of each dispatching type to obtain the management calibration dispatching task; dispatching the motorcycle based on the management calibration dispatching task.

[0063] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0064] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the motorcycle scheduling method provided by the above-mentioned methods, which method includes: obtaining multi-dimensional operation data of motorcycle stations; applying the multi-dimensional operation data based on a scheduling task prediction model to predict the initial scheduling task of the motorcycle station, and the scheduling task prediction model is constructed based on a machine learning algorithm; based on the predicted order turnover data of the motorcycle station, weighting the execution priority of the initial scheduling task to obtain a revenue calibration scheduling task; based on the scheduling parameters of the historical scheduling tasks of each scheduling type, weighting the execution priority of the revenue calibration scheduling tasks of each scheduling type to obtain a management calibration scheduling task; scheduling motorcycles based on the management calibration scheduling task.

[0065] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the motorcycle dispatching method provided by the above-mentioned methods, the method comprising: obtaining multi-dimensional operation data of motorcycle stations; applying the multi-dimensional operation data based on a dispatching task prediction model to predict an initial dispatching task of the motorcycle station, the dispatching task prediction model being constructed based on a machine learning algorithm; weighting the execution priority of the initial dispatching task based on the predicted order turnover data of the motorcycle station to obtain a revenue calibration dispatching task; weighting the execution priority of the revenue calibration dispatching task of each dispatching type based on the dispatching parameters of the historical dispatching tasks of each dispatching type to obtain a management calibration dispatching task; and dispatching motorcycles based on the management calibration dispatching task.

[0066] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0067] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0068] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for dispatching electric motorcycles, characterized in that: include: Obtain multi-dimensional operational data of motorcycle stations; Applying the multi-dimensional operation data based on a scheduling task prediction model to predict the initial scheduling task of the motorcycle station, wherein the scheduling task prediction model is constructed based on a machine learning algorithm; Based on the predicted order turnover data of the motorcycle station, weighting the execution priority of the initial scheduling task to obtain a revenue calibration scheduling task; Based on the scheduling parameters of the historical scheduling tasks of each scheduling type, weighting the execution priority of the revenue calibration scheduling tasks of each scheduling type, to obtain the management calibration scheduling task; The motorcycles are dispatched based on the management calibration scheduling task.

2. The method for dispatching motorcycles according to claim 1, characterized in that: The scheduling parameters of the historical scheduling tasks of each scheduling type are weighted to obtain the management calibration scheduling tasks, including: Based on the scheduling parameters of the historical scheduling tasks of each scheduling type, a single-vehicle scheduling route between a popular vehicle-deficient station and an unpopular station is obtained; Fitting the dispatching start and end points of the bicycle dispatching routes of each dispatching type to obtain the fitted dispatching routes of each dispatching type; Based on the scheduling efficiency of the fitting scheduling route of each scheduling type, the execution priority of the revenue calibration scheduling task of each scheduling type is weighted to obtain the management calibration scheduling task.

3. The method for dispatching motorcycles according to claim 2, characterized in that: The scheduling type includes at least one of a revenue task, an inspection task, an operation task, and a safety management task.

4. The method for dispatching electric motorcycles according to any one of claims 1 to 3, characterized in that: The step of weighting the execution priority of the initial scheduling task based on the predicted order turnover data of the motorcycle station to obtain a revenue calibration scheduling task includes: Determine high-yield stations based on the predicted order turnover data of the motorcycle stations; The execution priorities of the initial scheduling tasks corresponding to the high-yield sites are weighted to obtain the benefit calibration scheduling tasks.

5. The dispatching method for electric motorcycles according to any one of claims 1 to 3, characterized in that: The prediction parameters of the scheduling task prediction model include at least one of user riding demand at the station, scheduling trigger conditions, generation task time of the station, station priority, and real-time supply and demand conditions of the station; The prediction parameters of the scheduling task prediction model are determined based on the scheduling type.

6. The method for dispatching motorcycles according to claim 5, characterized in that: After the motorcycle is dispatched based on the management calibration dispatch task, the method further comprises: Obtaining dispatch effect data of the motorcycle station, wherein the dispatch effect data includes the first order generation time of the motorcycle station and / or the predicted order turnover data; Based on the scheduling effect data, the prediction parameters are optimized.

7. A dispatching device for an electric motorcycle, characterized in that: include: An acquisition unit, for acquiring multi-dimensional operation data of an electric motorcycle station; A prediction unit, applying the multi-dimensional operation data based on a scheduling task prediction model to predict an initial scheduling task for the motorcycle station, wherein the scheduling task prediction model is constructed based on a machine learning algorithm; A revenue calibration unit, based on the predicted order turnover data of the motorcycle station, weights the execution priority of the initial scheduling task to obtain a revenue calibration scheduling task; A management calibration unit, based on the scheduling parameters of the historical scheduling tasks of each scheduling type, weights the execution priority of the revenue calibration scheduling tasks of each scheduling type to obtain a management calibration scheduling task; The task dispatching unit dispatches the motorcycles based on the management calibration scheduling tasks.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the dispatching method for electric motorcycles as described in any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the dispatching method for electric motorcycles as described in any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the dispatching method for electric motorcycles as described in any one of claims 1 to 6 is implemented.

Images