Real-time analysis and prediction of ride-hailing fleet

Abstract

The application relates to the technical field of data processing, in particular to a real-time analysis and prediction method, system and device for online car-hailing capacity, which comprises the following steps: obtaining order data of a passenger side and point reporting data of a driver side; dividing the order data and the point reporting data in continuous time slices based on a preset time length; calculating the order quantity and the order acceptance quantity based on the order data of a current time slice; dividing the order data of the current time slice in a target number of continuous time windows, and calculating the quantity of non-fast-matched orders, the quantity of externally fast-matched orders and the quantity of internally fast-matched orders; determining the driver capacity based on the point reporting data of the current time slice; and determining online car-hailing real-time capacity matching information and predicting online car-hailing future capacity matching information based on the order index and the driver capacity. In this way, the dynamic change between passenger car demand and driver capacity can be analyzed in real time and accurately and comprehensively, so that a business reminder can be given to the driver, and the business can be ensured to be carried out efficiently and smoothly.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a real-time analysis and prediction method, system and device for online car-hailing capacity. BACKGROUND

[0002] In the online car-hailing business scenario, understanding the dynamic changes between passenger car demand and driver capacity can accurately remind the driver of the business, so as to ensure that the passenger car demand is met while ensuring the amount of orders processed by the driver. However, in the current technology, it is not possible to understand and analyze the dynamic changes between passenger car demand and driver capacity in real time and accurately and comprehensively. SUMMARY

[0003] Therefore, the purpose of the present application is to provide a real-time analysis and prediction method, system and device for online car-hailing capacity to overcome the problem that the dynamic changes between passenger car demand and driver capacity cannot be understood and analyzed in real time and accurately and comprehensively.

[0004] To achieve the above purpose, the present application adopts the following technical solutions:

[0005] In a first aspect, the present application provides a real-time analysis and prediction method for online car-hailing capacity, comprising:

[0006] Obtaining order data on the passenger side and report point data on the driver side;

[0007] Based on a preset time length, the order data and the report point data are divided into consecutive time slices;

[0008] Based on the order data of the current time slice, the order quantity and the order quantity are calculated; the order data of the current time slice is divided into a target number of consecutive time windows, and the number of non-fast matching orders, the number of external fast matching orders and the number of internal fast matching orders are calculated; wherein the number of non-fast matching orders is the number of orders whose order time and order or cancellation time are not in the same time window, the number of external fast matching orders is the number of orders whose order time and cancellation time are in the same time window, and the number of internal fast matching orders is the number of orders whose order time and order time are in the same time window;

[0009] Based on the report point data of the current time slice, the driver capacity is determined;

[0010] Based on the order index and the driver capacity of the same time slice, the real-time capacity matching information of online car-hailing is determined; wherein the order index includes the order quantity, the order quantity, the non-fast matching order quantity, the external fast matching order quantity and the internal fast matching order quantity;

[0011] Based on the order index and the driver capacity of a plurality of continuous time slices or time windows, the future capacity matching information of the online car-hailing is predicted.

[0012] Further, in some embodiments of the present application, further comprising: determining the corresponding cell information based on the location information in the order data and the report point data;

[0013] The order data of the current time slice includes: based on the order data of the current time slice, determining the order quantity and the order acceptance quantity for each city, cell and order level, respectively;

[0014] The order data of the current time slice is divided into a target number of continuous time windows, and the unfast matching order quantity, the external fast matching order quantity and the internal fast matching order quantity are calculated, including: dividing the order data of the current time slice into a target number of continuous time windows, and calculating the unfast matching order quantity, the external fast matching order quantity and the internal fast matching order quantity for each city, cell and order level, respectively.

[0015] Further, in some embodiments of the present application, before the order data and the report point data are divided into continuous time slices based on the preset time length, further comprising:

[0016] Based on the target field in the order data, a de-duplication identifier is generated for each order splicing, respectively; wherein the target field includes passenger ID, time slice, starting point and terminal coordinates;

[0017] For a plurality of orders with the same de-duplication identifier, only the data of one order is retained.

[0018] Further, in some embodiments of the present application, the driver capacity is determined based on the report point data of the current time slice, including:

[0019] The report point data of the current time slice is divided into a target number of continuous time windows, and the report point data in the time window is accumulated as the driver capacity corresponding to the time window; wherein the time length corresponding to the time window is the same as the time interval between two times of reporting the report point data by the driver.

[0020] Further, in some embodiments of the present application, the order data of the passenger side and the report point data of the driver side are obtained, including:

[0021] Collecting the order data of the passenger side and storing it in the preset basic database;

[0022] Collecting the report point data of the driver side and storing it in the preset message queue;

[0023] The order data is acquired from the basic database through a cloud transmission service, and the report point data is acquired from the message queue through a Flink computing engine and accessed to a big data real-time warehouse in a data stream manner for storage;

[0024] The data stored in the big data real-time warehouse is processed in real time through the Flink computing engine, and corresponding honeycomb information is acquired to build a data detail layer in the big data real-time warehouse, and the data after real-time processing is stored through the built data detail layer; the real-time processing includes parsing, filtering and integrating;

[0025] The data stored in the data detail layer is synchronized to the Doris database through the Flink computing engine, wherein the data synchronized to the Doris database is stored in the form of a data table.

[0026] Further, in some embodiments of the present application, further comprising:

[0027] Based on the real-time operation matching information and the future operation matching information of the online car-hailing, a vehicle guidance prompt is generated;

[0028] The vehicle guidance prompt includes a driving prompt for guiding the vehicle to drive, and a service prompt for guiding the vehicle to change the service level.

[0029] Further, in some embodiments of the present application, the real-time operation matching information of the online car-hailing is determined based on the order index and the driver operation capacity of the same time slice, comprising:

[0030] Based on the order quantity, the order acceptance quantity and the driver operation capacity, the matching ratio information of the real-time operation of the online car-hailing is determined;

[0031] Based on the non-fast matching order quantity, the external fast matching order quantity and the internal fast matching order quantity, the matching efficiency information of the real-time operation of the online car-hailing is calculated.

[0032] Further, in some embodiments of the present application,

[0033] The matching ratio information includes: supply-demand ratio and order matching rate; wherein the supply-demand ratio is the ratio of the order quantity to the driver operation capacity in a time slice, and the order matching rate is the ratio of the order acceptance quantity to the order quantity in a time slice;

[0034] The matching efficiency information includes: non-fast matching proportion, internal fast matching proportion and external fast matching proportion; wherein the non-fast matching proportion is the ratio of the non-fast matching order quantity to the order quantity in the corresponding time window, the internal fast matching proportion is the ratio of the internal fast matching order quantity to the order quantity in the corresponding time window, and the external fast matching proportion is the ratio of the external fast matching order quantity to the order quantity in the corresponding time window.

[0035] Secondly, this application provides a real-time analysis and prediction system for ride-hailing capacity, including:

[0036] The acquisition module is used to acquire order data from passengers and trip data from drivers.

[0037] The segmentation module is used to divide the order data and the reporting data into consecutive time slices based on a preset time length;

[0038] The indicator calculation module is used to calculate the number of orders placed and accepted based on the order data of the current time slice; it divides the order data of the current time slice into a target number of consecutive time windows, and calculates the number of orders that are not quickly matched, the number of externally quickly matched orders, and the number of internally quickly matched orders; where the number of orders that are not quickly matched is the number of orders whose order placement time and order acceptance or cancellation time are not in the same time window, the number of externally quickly matched orders is the number of orders whose order placement time and cancellation time are in the same time window, and the number of internally quickly matched orders is the number of orders whose order placement time and order acceptance time are in the same time window; and it determines driver capacity based on the reported data of the current time slice.

[0039] The capacity matching calculation module is used to determine the real-time capacity matching information of ride-hailing vehicles based on order indicators and driver capacity in the same time slice; wherein, the order indicators include the number of orders placed, the number of orders accepted, the number of orders not quickly matched, the number of externally quickly matched orders, and the number of internally quickly matched orders; and to predict the future capacity matching information of ride-hailing vehicles based on order indicators and driver capacity in multiple consecutive time slices or time windows.

[0040] Thirdly, this application provides a real-time analysis and prediction device for ride-hailing capacity, including a processor and a memory, wherein the processor is connected to the memory:

[0041] The processor is used to call and execute the program stored in the memory;

[0042] The memory is used to store the program, which is at least used in the aforementioned real-time analysis and prediction method for ride-hailing capacity.

[0043] This invention relates to the field of data processing technology, specifically to a method, system, and device for real-time analysis and prediction of ride-hailing capacity. The method includes: acquiring order data from passengers and dispatch data from drivers; dividing the order data and dispatch data into consecutive time slices based on a preset time length; calculating the number of orders placed and accepted based on the order data of the current time slice; dividing the order data of the current time slice into a target number of consecutive time windows, and calculating the number of orders not quickly matched, the number of externally quickly matched orders, and the number of internally quickly matched orders; wherein, the number of orders not quickly matched is the number of orders whose order placement time and order acceptance or cancellation time are not within the same time window, the number of externally quickly matched orders is the number of orders whose order placement time and cancellation time are within the same time window, and the number of internally quickly matched orders is the number of orders whose order placement time and order acceptance time are within the same time window; determining driver capacity based on the dispatch data of the current time slice; determining real-time ride-hailing capacity matching information based on order metrics and driver capacity in the same time slice; and predicting future ride-hailing capacity matching information based on order metrics and driver capacity in multiple consecutive time slices or time windows. In this way, the dynamic changes between passenger demand and driver capacity can be analyzed in real time, accurately and comprehensively, thereby providing drivers with business reminders and ensuring efficient and smooth operation. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is a flowchart illustrating the real-time analysis and prediction method for ride-hailing capacity provided in this embodiment of the invention.

[0046] Figure 2 This is a schematic diagram of the structure of the real-time analysis and prediction system for ride-hailing capacity provided in an embodiment of the present invention;

[0047] Figure 3 This is a schematic diagram of the structure of the real-time analysis and prediction device for ride-hailing capacity provided in an embodiment of the present invention. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be described in detail below. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other implementation methods obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0049] Figure 1 This is a flowchart illustrating the real-time analysis and prediction method for ride-hailing capacity provided in this embodiment of the invention. Please refer to [link / reference]. Figure 1 This embodiment may include the following steps:

[0050] S101. Obtain order data from the passenger terminal and trip report data from the driver terminal.

[0051] Specifically, the real-time analysis and prediction method for ride-hailing capacity provided in this application can acquire the aforementioned data using a pre-set database and message queue during the data acquisition phase. Then, the data is collected from the database and message queue through cloud transmission services and the Flink computing engine, and then connected to the constructed big data real-time warehouse in the form of a data stream. The Flink computing engine is then used to process the data in the big data real-time warehouse, such as parsing, filtering, and integrating, and the processed data is synchronized to the Doris database and stored in the form of a data table.

[0052] S102. Based on a preset time length, divide the order data and reporting data into consecutive time slices.

[0053] Specifically, the preset time length can be 10 minutes, and the data table processed by the above steps is divided into time slices every 10 minutes.

[0054] S103. Based on the order data of the current time slice, calculate the number of orders placed and the number of orders accepted; divide the order data of the current time slice into a target number of consecutive time windows, and calculate the number of orders that are not quickly matched, the number of externally quickly matched orders, and the number of internally quickly matched orders.

[0055] The number of orders that were not quickly matched refers to the number of orders whose order placement time and order acceptance or cancellation time were not within the same time window. The number of externally quickly matched orders refers to the number of orders whose order placement time and cancellation time were within the same time window. The number of internally quickly matched orders refers to the number of orders whose order placement time and order acceptance time were within the same time window.

[0056] It should be noted that, among the above order metrics, externally matched orders are those that originate from other platforms or systems (not the same as the real-time analysis and prediction system for ride-hailing capacity used to execute the real-time analysis and prediction method for ride-hailing capacity in this application) and are subsequently cancelled within a short period of time. In such cases, the order can be considered to have been processed by the external platform itself, i.e., externally matched orders.

[0057] S104. Determine driver capacity based on the reported data of the current time slice.

[0058] The reported data refers to the data reported by the driver to display their geographical location.

[0059] S105. Based on order metrics and driver capacity within the same time slice, determine real-time ride-hailing capacity matching information.

[0060] S106. Based on order metrics and driver capacity from multiple consecutive time slices or time windows, predict future ride-hailing capacity matching information.

[0061] The order metrics include the number of orders placed, the number of orders accepted, the number of orders not quickly matched, the number of externally quickly matched orders, and the number of internally quickly matched orders. The specific prediction method involves calculating and analyzing the trends of each of these order metrics using known values ​​from multiple consecutive time periods. Based on this trend analysis, the corresponding future metric values ​​can be obtained (the same applies to driver capacity), and then the corresponding capacity matching information can be calculated. This analysis can be performed using precise mathematical calculations or by utilizing a pre-defined model.

[0062] In addition, in some other embodiments of this application, the changing trends of real-time capacity matching information of multiple consecutive time slices can be directly analyzed to directly predict the capacity matching information for future times.

[0063] The real-time analysis and prediction method for ride-hailing capacity provided in this application obtains order data from passengers and dispatch data from drivers through the aforementioned means; divides the order data and dispatch data into consecutive time slices based on a preset time length; calculates the number of orders placed and accepted based on the order data of the current time slice; divides the order data of the current time slice into a target number of consecutive time windows and calculates the number of orders not quickly matched, the number of externally quickly matched orders, and the number of internally quickly matched orders; determines driver capacity based on the dispatch data of the current time slice; determines real-time ride-hailing capacity matching information based on order indicators and driver capacity in the same time slice; and predicts future ride-hailing capacity matching information based on order indicators and driver capacity in multiple consecutive time slices or time windows. This method can analyze the dynamic changes between passenger demand and driver capacity in real time, accurately and comprehensively, thereby providing business reminders to drivers and ensuring efficient and smooth operation of the business.

[0064] Furthermore, in some embodiments of this application, obtaining passenger order data and driver reporting data may include: collecting passenger order data and storing it in a preset basic database; collecting driver reporting data and storing it in a preset message queue; obtaining order data from the basic database via cloud transmission service and obtaining reporting data from the message queue via Flink computing engine, and connecting them to a big data real-time warehouse for storage in the form of data streams; processing the data stored in the big data real-time warehouse in real time via Flink computing engine and obtaining corresponding cellular information to build a data detail layer in the big data real-time warehouse, and storing the real-time processed data through the constructed data detail layer; synchronizing the data stored in the data detail layer to the Doris database via Flink computing engine, wherein the data synchronized to the Doris database is stored in the form of data tables.

[0065] Specifically, for example, passenger order data is stored in MySQL, and driver trip data is uploaded and stored in the RabbitMQ message queue. Then, cloud transport services such as Alibaba Cloud Transport Service are used to collect order data from MySQL, and Flink computing engine is used to collect trip data from RabbitMQ. Finally, both data streams are connected to the Operational Data Store (ODS) layer of the Kafka big data real-time warehouse for storage.

[0066] Building upon this foundation, the Flink computing engine is used to process the data from the aforementioned ODS layer in real time, including data parsing, filtering, and integration. Furthermore, by invoking pre-defined algorithms such as the UberH3 algorithm, corresponding warehouse information is calculated and retrieved based on the location coordinates of order and reporting data. Finally, a Data Warehouse Detail (DWD) layer is constructed to support subsequent data analysis.

[0067] Based on this, the order data and dispatch data are synchronized to the Doris data table in real time through the Flink computing engine, including synchronizing the order data to the order table and synchronizing the dispatch data to the capacity index aggregation.

[0068] Specifically, in this application, the order table adopts a unique key model design. The unique key definition information in the table includes order placement time, city ID, Hive ID, and order ID. Key indicator fields include order level, order acceptance time, cancellation time, and deduplication ID.

[0069] It should be noted that, in this embodiment of the application, to ensure accuracy when calculating real-time order metrics, the order data must be deduplicated. The deduplication process specifically includes: generating a deduplication identifier, i.e., a deduplication ID, for each order based on the target fields in the order data; wherein, the target fields include passenger ID, the time slice to which the order was placed, and the coordinates of the origin and destination; and for multiple orders with the same deduplication identifier, only the data of one order is retained.

[0070] In practical applications, the Doris database can leverage the principle of scheduled tasks. For example, a scheduled task system can be used within the Doris database to launch tasks after a 10-second delay every ten-minute period. This allows for the calculation and analysis of various order metrics.

[0071] Specifically, this includes, for order volume, counting the number of duplicate orders placed within the ten minutes prior to the current time; and for order acceptance volume, counting the number of duplicate orders accepted within the ten minutes prior to the current time.

[0072] Furthermore, for the current time slice, multiple boundaries are defined to divide the time slice into multiple time windows, such as 30 seconds, to calculate the number of orders that are not quickly matched, the number of externally quickly matched orders, and the number of internally quickly matched orders. For orders that are not quickly matched, the number of orders whose order placement time is within a certain 30-second window but whose order acceptance or cancellation time is not within that window is counted. For externally quickly matched orders, the number of orders whose order placement time and cancellation time are both within a 30-second window is counted. For internally quickly matched orders, the number of orders whose order placement time and order acceptance time are both within a 30-second window is counted.

[0073] In the above process, calculation and analysis are performed by delaying the time, that is, the task will start after a 10-second delay after each ten-minute cycle, which can ensure the integrity of the data; at the same time, by dividing the ten minutes into 30-second windows, real-time data statistics and analysis can be achieved in a fine-grained manner; and by generating deduplication IDs to deduplicate order data, the accuracy of statistical results can be ensured.

[0074] Furthermore, in this embodiment of the application, determining driver capacity based on the reporting data of the current time slice includes: dividing the reporting data of the current time slice into a target number of consecutive time windows, and accumulating the reporting data within the time windows as the driver capacity corresponding to the time windows; wherein, the duration corresponding to the time window is the same as the time interval between two reports of reporting data by the driver.

[0075] Specifically, as mentioned above, the Flink computing engine can be used to synchronize driver-reported trip data to the Doris capacity index aggregation table in real time. This table uses Doris's aggregation model for data aggregation. The aggregation key of the table consists of the following fields: city, time slice, Hive ID, vehicle type, time, and time window information. Based on this, the sum of the number of drivers accepting orders is taken as the driver's capacity. It should be noted that in practical applications, the duration of the time window can be the same as the time interval between two driver-reported trip data entries. For example, if a driver reports one trip data entry every 30 seconds, the time window should also be set to 30 seconds. In this way, when calculating driver capacity based on the cumulative time window, no additional deduplication operation is required.

[0076] Based on this, the calculation of order volume and order acceptance volume based on the order data of the current time slice mentioned in the above embodiments includes: determining the order volume and order acceptance volume for each city, honeycomb, and order level based on the order data of the current time slice.

[0077] Specifically, the matching ratio of real-time ride-hailing capacity can be determined based on the number of orders placed, the number of orders accepted, and driver capacity; and the matching efficiency of real-time ride-hailing capacity can be calculated based on the number of orders that are not quickly matched, the number of externally quickly matched orders, and the number of internally quickly matched orders.

[0078] The matching ratio information includes the supply-demand ratio and the order matching rate. The supply-demand ratio is the ratio of the number of orders placed in a time slot to the driver's capacity, and the order matching rate is the ratio of the number of orders accepted in a time slot to the number of orders placed.

[0079] Matching efficiency information includes the percentage of orders not matched quickly, the percentage of internal quick matches, and the percentage of external quick matches. The percentage of orders not matched quickly is the ratio of the number of orders not matched quickly to the number of orders placed in the corresponding time window. The percentage of internal quick matches is the ratio of the number of internal quick matches to the number of orders placed in the corresponding time window. The percentage of external quick matches is the ratio of the number of external quick matches to the number of orders placed in the corresponding time window.

[0080] The above embodiment mentions dividing the order data of the current time slice into a target number of consecutive time windows and calculating the number of non-fast-matched orders, external fast-matched orders, and internal fast-matched orders, which includes: dividing the order data of the current time slice into a target number of consecutive time windows, and calculating the number of non-fast-matched orders, external fast-matched orders, and internal fast-matched orders for each city, Hive, and order level.

[0081] It should be noted that the above analysis and calculation process in this application can be performed separately for different dimensions, such as determining the order volume and order acceptance volume for each city, honeycomb network, and order level. Furthermore, the number of orders not quickly matched, externally quickly matched, and internally quickly matched orders can be calculated separately for each city, honeycomb network, and order level.

[0082] Based on this, vehicle guidance prompts can be generated using real-time and future ride-hailing capacity matching information. These prompts include driving prompts to guide the vehicle's movement and service prompts to guide the vehicle to change its service level (e.g., prompting vehicles offering luxury service to lower their service level to increase the number of completed orders, such as by reducing to comfort or economy service).

[0083] For example, in some embodiments, when the real-time ride-hailing capacity matching information shows that the supply and demand of a certain hive (region) is greater than or equal to a preset judgment threshold, and when the ride-hailing capacity matching information predicts that this state may continue for the next preset number of minutes (multiple conditions can be set simultaneously using multiple pieces of information such as the calculated supply-demand ratio and order matching rate), a driving prompt can be pushed to drivers within a target kilometer of that hive who currently have no orders: "Currently, there is a high demand for orders in the XX area; it is recommended to proceed to the specific route."; and when the proportion of quick matching within "luxury" orders in a certain area is less than or equal to a preset judgment threshold, and the supply-demand ratio of "economy" orders is greater than or equal to a preset judgment threshold, a prompt can be pushed to luxury drivers in that area: "Currently, there is a high demand for economy orders; it is recommended to temporarily switch the service level to economy."

[0084] The real-time analysis and prediction method for ride-hailing capacity provided by this invention combines spatial and temporal dimensions and uses multiple continuous time segments for data analysis and comparison. This not only enables real-time analysis in a shorter time, significantly reducing computational latency, but also more sensitively and accurately determines the relationship between passenger demand and ride-hailing capacity. Simultaneously, by utilizing the Flink computing engine, real-time data processing is performed on data from multiple data sources, and the data is synchronized to a Doris database for storage in the form of data tables. The data tables are then used for computational analysis, ensuring the consistency of multiple data streams at a given moment.

[0085] Based on the same inventive concept, this application also provides a real-time analysis and prediction system for ride-hailing capacity, used to implement the above-described method embodiments. Figure 2 This is a schematic diagram of the structure of the real-time analysis and prediction system for ride-hailing capacity provided in an embodiment of the present invention, as shown below. Figure 2 As shown, the system includes:

[0086] The acquisition module 11 is used to acquire order data from the passenger end and trip data from the driver end.

[0087] The segmentation module 12 is used to divide order data and reporting data into consecutive time slices based on a preset time length.

[0088] The indicator calculation module 13 is used to calculate the number of orders placed and accepted based on the order data of the current time slice; divide the order data of the current time slice into a target number of consecutive time windows, and calculate the number of orders that are not quickly matched, the number of externally quickly matched orders, and the number of internally quickly matched orders; wherein, the number of orders that are not quickly matched is the number of orders whose order placement time and order acceptance or cancellation time are not in the same time window, the number of externally quickly matched orders is the number of orders whose order placement time and cancellation time are in the same time window, and the number of internally quickly matched orders is the number of orders whose order placement time and order acceptance time are in the same time window; and determine the driver capacity based on the reporting data of the current time slice.

[0089] The capacity matching calculation module 14 is used to determine the real-time capacity matching information of ride-hailing vehicles based on order indicators and driver capacity in the same time slice; wherein, the order indicators include the number of orders placed, the number of orders accepted, the number of orders not quickly matched, the number of externally quickly matched orders, and the number of internally quickly matched orders; and to predict the future capacity matching information of ride-hailing vehicles based on order indicators and driver capacity in multiple consecutive time slices or time windows.

[0090] Regarding the system in the above embodiments, the specific manner in which each module performs its operations has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0091] This invention also provides a real-time analysis and prediction device for ride-hailing capacity, used to implement the above-described method embodiments. Figure 3 This is a schematic diagram of the structure of the real-time analysis and prediction device for ride-hailing capacity provided in an embodiment of the present invention, as shown below. Figure 3 As shown, the real-time analysis and prediction device for ride-hailing capacity in this embodiment includes a processor 21 and a memory 22, with the processor 21 connected to the memory 22. The processor 21 is used to call and execute programs stored in the memory 22; the memory 22 is used to store programs, which are at least used to execute the real-time analysis and prediction method for ride-hailing capacity in the above embodiments.

[0092] The specific implementation scheme of the real-time analysis and prediction device for ride-hailing capacity provided in this application embodiment can refer to the implementation scheme of the real-time analysis and prediction method for ride-hailing capacity in any of the above embodiments, and will not be repeated here.

[0093] It is understood that the same or similar parts in the above embodiments can be referred to each other, and the contents not described in detail in some embodiments can be referred to the same or similar contents in other embodiments.

[0094] It should be noted that in the description of this invention, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, in the description of this invention, unless otherwise stated, "a plurality of" means at least two.

[0095] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of the invention pertain.

[0096] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0097] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it includes one or a combination of the steps of the method embodiments.

[0098] Furthermore, the functional units in the various embodiments of the present invention can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0099] The storage media mentioned above can be read-only memory, disk, or optical disk, etc.

[0100] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0101] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for real-time analysis and prediction of ride-hailing capacity, characterized in that, include: Obtain order data from passengers and trip data from drivers; Based on a preset time length, the order data and the reporting data are divided into consecutive time slices; Based on the order data of the current time slice, calculate the number of orders placed and accepted; divide the order data of the current time slice into a target number of consecutive time windows, and calculate the number of orders not quickly matched, the number of externally quickly matched orders, and the number of internally quickly matched orders; where the number of orders not quickly matched is the number of orders whose order placement time and order acceptance or cancellation time are not in the same time window, the number of externally quickly matched orders is the number of orders whose order placement time and cancellation time are in the same time window, and the number of internally quickly matched orders is the number of orders whose order placement time and order acceptance time are in the same time window; Determine driver capacity based on the reported data for the current time slice; Based on order metrics and driver capacity within the same time slice, real-time ride-hailing capacity matching information is determined; wherein, the order metrics include the number of orders placed, the number of orders accepted, the number of orders not quickly matched, the number of externally quickly matched orders, and the number of internally quickly matched orders; Based on order metrics and driver capacity across multiple consecutive time slices or time windows, predict future ride-hailing capacity matching information; The determination of real-time ride-hailing capacity matching information based on order metrics and driver capacity within the same time slice includes: Based on the number of orders placed, the number of orders accepted, and driver capacity, determine the matching ratio information of real-time ride-hailing capacity; Based on the number of non-fast-matched orders, the number of externally fast-matched orders, and the number of internally fast-matched orders, calculate the matching efficiency information of real-time ride-hailing capacity. The matching ratio information includes: supply-demand ratio and order matching rate; wherein, the supply-demand ratio is the ratio of the number of orders placed in a time slot to the driver's capacity, and the order matching rate is the ratio of the number of orders accepted in a time slot to the number of orders placed. The matching efficiency information includes: the percentage of orders not matched quickly, the percentage of internal quick matches, and the percentage of external quick matches; wherein, the percentage of orders not matched quickly is the ratio of the number of orders not matched quickly to the number of orders placed in the corresponding time window, the percentage of internal quick matches is the ratio of the number of internal quick matches to the number of orders placed in the corresponding time window, and the percentage of external quick matches is the ratio of the number of external quick matches to the number of orders placed in the corresponding time window.

2. The method for real-time analysis and prediction of ride-hailing capacity according to claim 1, characterized in that, Also includes: Based on the location information in the order data and the reporting data, the corresponding hive information is determined; The calculation of order volume and order acceptance volume based on the order data of the current time slice includes: determining the order volume and order acceptance volume for each city, honeycomb, and order level based on the order data of the current time slice; The step of dividing the order data of the current time slice into a target number of consecutive time windows and calculating the number of non-fast-matched orders, external fast-matched orders, and internal fast-matched orders includes: dividing the order data of the current time slice into a target number of consecutive time windows, and calculating the number of non-fast-matched orders, external fast-matched orders, and internal fast-matched orders for each city, Hive, and order level.

3. The method for real-time analysis and prediction of ride-hailing capacity according to claim 1, characterized in that, Before dividing the order data and the reporting data into consecutive time slices based on a preset time length, the method further includes: Based on the target fields in the order data, a deduplication identifier is generated for each order; wherein, the target fields include passenger ID, time slice, origin and destination coordinates; For multiple orders with the same deduplication identifier, only the data of one order is retained.

4. The method for real-time analysis and prediction of ride-hailing capacity according to claim 1, characterized in that, The determination of driver capacity based on the reporting data of the current time slice includes: The reporting data of the current time slice is divided into a target number of consecutive time windows, and the reporting data within each time window is accumulated as the driver capacity corresponding to the time window; wherein, the duration of the time window is the same as the time interval between two driver reports of reporting data.

5. The method for real-time analysis and prediction of ride-hailing capacity according to claim 1, characterized in that, The acquisition of order data from the passenger terminal and trip report data from the driver terminal includes: Collect passenger order data and store it in a pre-set basic database; Collect the location data reported by the driver and store it in a preset message queue; The order data is obtained from the basic database through cloud transmission service, and the reporting data is obtained from the message queue through Flink computing engine, and then connected to the big data real-time warehouse for storage in the form of data stream; The Flink computing engine is used to process the data stored in the big data real-time warehouse in real time and obtain the corresponding hive information to build a data detail layer in the big data real-time warehouse. The real-time processed data is stored in the constructed data detail layer. The real-time processing includes parsing, filtering and integration. The data stored in the data detail layer is synchronized to the Doris database using the Flink computing engine. The data synchronized to the Doris database is stored in the form of data tables.

6. The method for real-time analysis and prediction of ride-hailing capacity according to claim 1, characterized in that, Also includes: Based on real-time and future ride-hailing capacity matching information, vehicle guidance prompts are generated. The vehicle guidance prompts include driving prompts to guide the vehicle to travel, and service prompts to guide the vehicle to change its service level.

7. A real-time analysis and prediction system for ride-hailing capacity, used to execute the method according to any one of claims 1-6, characterized in that, include: The acquisition module is used to acquire order data from passengers and trip data from drivers. The segmentation module is used to divide the order data and the reporting data into consecutive time slices based on a preset time length; The indicator calculation module is used to calculate the number of orders placed and accepted based on the order data of the current time slice; it divides the order data of the current time slice into a target number of consecutive time windows, and calculates the number of orders that are not quickly matched, the number of externally quickly matched orders, and the number of internally quickly matched orders; where the number of orders that are not quickly matched is the number of orders whose order placement time and order acceptance or cancellation time are not in the same time window, the number of externally quickly matched orders is the number of orders whose order placement time and cancellation time are in the same time window, and the number of internally quickly matched orders is the number of orders whose order placement time and order acceptance time are in the same time window; and it determines driver capacity based on the reported data of the current time slice. The capacity matching calculation module is used to determine the real-time capacity matching information of ride-hailing vehicles based on order indicators and driver capacity in the same time slice; wherein, the order indicators include the number of orders placed, the number of orders accepted, the number of orders not quickly matched, the number of externally quickly matched orders, and the number of internally quickly matched orders; and to predict the future capacity matching information of ride-hailing vehicles based on order indicators and driver capacity in multiple consecutive time slices or time windows.

8. A real-time analysis and prediction device for ride-hailing capacity, characterized in that, It includes a processor and a memory, wherein the processor is connected to the memory: The processor is used to call and execute the program stored in the memory; The memory is used to store the program, which is at least used to execute the real-time analysis and prediction method for ride-hailing capacity as described in any one of claims 1-6.

Images