Online car-hailing operation data fluctuation analysis method and device, equipment and medium

Abstract

The invention relates to an online car-hailing operation data fluctuation analysis method and device, computer equipment and a storage medium. The method comprises the following steps: in response to an attribution analysis instruction, obtaining basic operation data of an online hailed car generated in a time period when a core operation index abnormally fluctuates; splitting the basic operation data according to different analysis paths to generate a plurality of groups of to-be-calculated operation data, inputting each group of to-be-calculated operation data into a corresponding fluctuation analysis model, and enabling each fluctuation analysis model to correspond to each analysis path; analyzing and calculating each group of to-be-calculated operation data through each fluctuation analysis model to obtain a path analysis result output by each fluctuation analysis model; and performing fluctuation attribution analysis according to each path analysis result to obtain a fluctuation reason analysis conclusion for the core operation index. By adopting the method, the analysis efficiency and accuracy of the online car-hailing operation data can be improved.

Description

Technical Field

[0001] The present application relates to the field of data analysis technology, and in particular to a method, device, computer equipment and storage medium for analyzing fluctuations in online ride-hailing operation data. Background Art

[0002] With the rapid development of online ride-hailing platforms, data analysis technology has been gradually applied to the analysis of online ride-hailing operation data. However, the platform faces the challenges of rapid changes in multi-channel passenger demand and complex management of diversified transport supply. The analysis of operational data generated during the operation process has also become more complicated due to the sharp increase in data volume and data dimensions.

[0003] Traditional analysis methods for online ride-hailing operation data have a low degree of automation and rely on manual observation. Therefore, they are unable to timely and effectively identify key issues or potential opportunities that affect data fluctuations during the operation process. Summary of the invention

[0004] Based on this, it is necessary to provide an analysis method, device, computer equipment and storage medium for online ride-hailing operation data fluctuations that can improve the efficiency and accuracy of online ride-hailing operation data analysis in response to the above-mentioned technical problems.

[0005] In a first aspect, a method for analyzing fluctuations in online car-hailing operation data is provided, the method comprising:

[0006] In response to the attribution analysis instruction, obtain the basic operation data of the online car-hailing service generated during the time period when the core operation indicators fluctuate abnormally;

[0007] The basic operation data is split according to different analysis paths to generate multiple groups of operation data to be calculated, and each group of operation data to be calculated is input into a corresponding fluctuation analysis model, and each fluctuation analysis model corresponds to each analysis path;

[0008] Each group of operation data to be calculated is analyzed and calculated by each fluctuation analysis model to obtain the path analysis results output by each fluctuation analysis model;

[0009] Based on the analysis results of each path, fluctuation attribution analysis is performed to obtain analysis conclusions on the causes of fluctuations in core operating indicators.

[0010] In some embodiments, the basic operation data is split according to different analysis paths to generate multiple groups of operation data to be calculated, including:

[0011] Determine the indicators to be calculated for each analysis path;

[0012] The basic operation data is split according to each indicator to be calculated to generate multiple groups of operation data to be calculated that match the calculation requirements of each indicator to be calculated.

[0013] In some embodiments, the analysis path includes a passenger-side demand path, a driver-side supply path, and an efficiency evaluation path.

[0014] In some embodiments, the indicators to be calculated for the demand path on the passenger side include at least one of the order volume, matching volume, response volume, order completion volume, cancellation volume and response rate; the supply path on the driver side includes at least one of the number of drivers dispatched, the number of drivers completing orders and the order completion rate; the efficiency evaluation path includes at least one indicator characterizing the driver's work quality.

[0015] In some embodiments, each group of to-be-calculated operational data is analyzed and calculated by each fluctuation analysis model to obtain the path analysis results output by each fluctuation analysis model, including:

[0016] Decompose the operational data to be calculated into different layers according to different layer dimensions;

[0017] Obtain historical operational data that is stratified by each stratification dimension;

[0018] Calculate the data fluctuation changes of each layer of the operating data to be calculated based on the historical operating data;

[0019] According to the data fluctuation changes of each stratum, the stratified groups whose fluctuation impact on the analysis path is greater than the preset threshold are determined.

[0020] In some embodiments, the hierarchical dimension includes at least one of a driver hierarchical dimension, a time period hierarchical dimension, and an order type hierarchical dimension.

[0021] In some embodiments, fluctuation attribution analysis is performed based on the analysis results of each path to obtain fluctuation cause analysis conclusions for core operating indicators, including:

[0022] Determine the factors affecting fluctuations, which include internal and external factors;

[0023] Collecting the historical real-time data of internal factors corresponding to each internal factor and the historical real-time data of external factors corresponding to each external factor;

[0024] Predict the fluctuation trend of core operating indicators under different fluctuation factors based on historical internal and external real-time data;

[0025] Compare the path analysis results with the fluctuation trends under each fluctuation influencing factor to determine the analysis conclusions on the fluctuation causes of core operating indicators.

[0026] In a second aspect, a device for analyzing fluctuations in online car-hailing operation data is provided, the device comprising:

[0027] A data acquisition module, for responding to the attribution analysis instruction and acquiring the basic operation data of the online car-hailing service generated during the period when the core operation indicators fluctuate abnormally;

[0028] A data splitting module is used to split the basic operation data according to different analysis paths to generate multiple groups of operation data to be calculated, and input each group of operation data to be calculated into a corresponding fluctuation analysis model, each fluctuation analysis model corresponds to each analysis path;

[0029] The split calculation module is used to analyze and calculate each group of operation data to be calculated through each fluctuation analysis model, and obtain the path analysis results output by each fluctuation analysis model;

[0030] The attribution hit module is used to perform fluctuation attribution analysis based on the analysis results of each path and obtain the analysis conclusions of the fluctuation causes for the core operating indicators.

[0031] According to a third aspect, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of a method for analyzing fluctuations in online car-hailing operation data when executing the computer program.

[0032] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps of a method for analyzing fluctuations in online car-hailing operation data are implemented.

[0033] The above-mentioned analysis method, device, computer equipment and storage medium for fluctuations in online ride-hailing operation data split the basic operation data according to different analysis paths to obtain segmented data in different directions, and perform fluctuation attribution analysis based on the path analysis results of each analysis path, so as to efficiently and accurately obtain analysis conclusions on the causes of fluctuations in online ride-hailing operation data for multiple channels and multiple supplies, thereby improving the analysis efficiency and accuracy of online ride-hailing operation data. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A diagram of an application environment of a method for analyzing fluctuations in online car-hailing operation data in some embodiments;

[0035] Figure 2 A flowchart of a method for analyzing fluctuations in online ride-hailing operation data in some embodiments;

[0036] Figure 3 An interactive time sequence diagram of a method for analyzing fluctuations in online car-hailing operation data in some embodiments;

[0037] Figure 4 It is a structural block diagram of a device for analyzing fluctuations in online car-hailing operation data in some embodiments;

[0038] Figure 5 1 is a diagram of the internal structure of a computer device in some embodiments. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0040] The analysis method for the fluctuation of online car-hailing operation data provided in this application can be applied to Figure 1 In the application environment shown, the terminal 102 communicates with the server 104 via the network.

[0041] In some embodiments, the terminal 102 may trigger an attribution analysis instruction, and the server 104 may respond to the attribution analysis instruction and obtain the basic operation data of the online car-hailing vehicle generated during the time period when the core operation indicators fluctuate abnormally, split the basic operation data according to different analysis paths to generate multiple groups of operation data to be calculated, input each group of operation data to be calculated into a corresponding fluctuation analysis model, and each fluctuation analysis model corresponds to each analysis path; analyze and calculate each group of operation data to be calculated by each fluctuation analysis model, and obtain the path analysis results output by each fluctuation analysis model; perform fluctuation attribution analysis based on the analysis results of each path, and obtain the fluctuation cause analysis conclusion for the core operation indicators. Furthermore, the fluctuation cause analysis conclusion may be assembled, integrated, or screened and then returned to the terminal 102.

[0042] The terminal 102 may be, but is not limited to, various personal computers, laptop computers, smart phones, tablet computers, and portable wearable devices, and the server 104 may be implemented as an independent server or a server cluster consisting of multiple servers.

[0043] In some embodiments, Figure 2 As shown in the figure, a method for analyzing the fluctuation of online car-hailing operation data is provided. Figure 1 The terminal in is used as an example to illustrate, including the following steps:

[0044] Step 202, in response to the attribution analysis instruction, obtain the basic operating data of the online car-hailing service generated during the time period when the core operating indicators experience abnormal fluctuations.

[0045] In this step, the analysis of fluctuations in online ride-hailing operation data, that is, after the attribution analysis instruction for fluctuations in online ride-hailing operation data is triggered, the attribution analysis process is started to obtain the basic operation data of the online ride-hailing generated during the period when the core operation indicators fluctuate abnormally for analysis.

[0046] For example, the trigger conditions can be divided into two modes: active trigger and passive trigger:

[0047] a. Active triggering can be triggered by the user of the terminal. By selecting the time point when the operating data fluctuates, the operating data between the specified two points in time can be analyzed.

[0048] b. Passive triggering can be triggered by daily scheduled triggering tasks, or automatically triggered when the triggering conditions are met. The two triggering modes correspond to the user's year-on-year needs and monitoring needs respectively.

[0049] Step 204, split the basic operation data according to different analysis paths to generate multiple groups of operation data to be calculated, input each group of operation data to be calculated into a corresponding fluctuation analysis model, and each fluctuation analysis model corresponds to each analysis path.

[0050] Among them, the analysis path can be divided according to the analysis direction of business needs. For example, in the online car-hailing industry, in order to improve the online car-hailing business with multiple channels and multiple supplies, the analysis path can conduct in-depth data decomposition based on the three directions of demand, supply, and efficiency.

[0051] For example, the analysis path may include a demand path on the passenger side, a supply path on the driver side, and an efficiency evaluation path. Different analysis paths may correspond to different indicators to be calculated.

[0052] In some embodiments, the basic operation data is split according to different analysis paths to generate multiple groups of operation data to be calculated, including: determining the indicators to be calculated for each analysis path; splitting the basic operation data according to each indicator to be calculated to generate multiple groups of operation data to be calculated that match the calculation requirements of each indicator to be calculated.

[0053] For example, the demand path on the passenger side can be decomposed into the order quantity, matching quantity, response quantity, order completion quantity, cancellation quantity, response rate and other related indicators to be calculated. The supply path on the driver side can be decomposed according to the capacity quantity, and decomposed into the number of drivers dispatched, the number of drivers completing orders, and the dispatch order completion rate to be calculated. The efficiency evaluation path can be mainly decomposed according to the quality of capacity, and decomposed into various indicators to be calculated that can represent the quality of the driver's work, such as the time of dispatching, service points, and the time of listening to orders at a fixed price.

[0054] By decomposing data based on analysis paths, the basic operating data can be divided into three directions: the demand path on the passenger side, the supply path on the driver side, and the efficiency evaluation path. The data is then suitable for calculating the indicators to be calculated in each path. This allows for orderly path division of complex and disorganized data, thereby improving the efficiency of data analysis.

[0055] Step 206 , analyzing and calculating each group of to-be-calculated operating data using each fluctuation analysis model, and obtaining path analysis results output by each fluctuation analysis model.

[0056] Specifically, each fluctuation analysis model corresponds to each analysis path, that is, each analysis path calls the corresponding fluctuation analysis model to analyze and calculate the corresponding operation data to be calculated that is decomposed into it, that is, calculates the fluctuation of the indicator to be calculated under the analysis path and performs analysis and judgment, so that different analysis paths output path analysis results in different directions (for example, demand on the passenger side, supply on the driver side, and efficiency evaluation).

[0057] Step 208, performing fluctuation attribution analysis based on the analysis results of each path, and obtaining fluctuation cause analysis conclusions for the core operating indicators.

[0058] In this step, the results of each path analysis can be integrated, and by analyzing the changing trends between each fluctuation influencing factor and the fluctuation of the core operating indicators in the historical data, it can be determined whether each path analysis result hits the changing trends between each fluctuation influencing factor and the fluctuation of the core operating indicators, so as to conduct attribution analysis, determine the fluctuation influencing factors that affect the fluctuation of the core operating indicators, and then obtain the analysis conclusion of the cause of the fluctuation.

[0059] The above-mentioned method for analyzing fluctuations in online ride-hailing operation data splits the basic operation data according to different analysis paths to obtain segmented data in different directions, and performs fluctuation attribution analysis based on the path analysis results of each analysis path, so as to efficiently and accurately obtain analysis conclusions on the causes of fluctuations in online ride-hailing operation data for multiple channels and multiple supplies, thereby improving the analysis efficiency and accuracy of online ride-hailing operation data.

[0060] In some embodiments, each fluctuation analysis model is used to analyze and calculate each group of operating data to be calculated, and path analysis results output by each fluctuation analysis model are obtained, including: decomposing the operating data to be calculated into layers according to different hierarchical dimensions; obtaining historical operating data that is hierarchically layered according to each hierarchical dimension; calculating the data fluctuation changes of each layer of the operating data to be calculated based on the historical operating data; and determining the hierarchical groups whose fluctuation impact on the analysis path is greater than a preset threshold based on the data fluctuation changes of each layer.

[0061] In this embodiment, the accuracy of the analysis can be improved by breaking down the to-be-calculated operational data of each analysis path into layers to drill down into the data.

[0062] Exemplarily, the data stratification is described in detail by taking the demand path on the passenger side, the supply path on the driver side, and the efficiency evaluation path as examples.

[0063] a. The demand path on the passenger side can be decomposed into order quantity, matching quantity, response quantity, order completion quantity, cancellation quantity, response rate and other calculation indicators according to the order dimension. Below, the order quantity is used as an example to illustrate: the analysis method for the order quantity indicator is to decompose the data in various dimensions and conduct detailed data exploration. For example, the data is decomposed in layers according to the time period dimension and the order type dimension (layer dimension), and the data fluctuation changes of each layer are obtained by comparing the historical data performance. The layer with higher data fluctuation ranking is the order layer group or time layer group that has the greatest impact on the analysis path. Then, by comparing the hit results of the attribution module for the hit time layer group, a reliable fluctuation cause analysis conclusion is obtained.

[0064] b. The supply path on the driver side can be mainly decomposed according to the number of transport capacity to obtain the number of drivers dispatched, the number of drivers who complete orders, the dispatch completion rate and other calculation indicators. Below, the number of drivers dispatched is used as an example to illustrate: the analysis method of the indicator of the number of drivers dispatched is to stratify the data into driver dimensions and time period dimensions (stratification dimensions) to conduct data exploration. By comparing historical data, it can also be further combined with the prediction of individual driver data to determine which driver stratification groups or which individual driver's behavioral problems have a greater impact on the data fluctuations of the analysis path, and then by performing a hit comparison with the attribution module, a reliable analysis conclusion on the cause of the fluctuation can be drawn.

[0065] c. The efficiency evaluation path can be mainly decomposed according to the quality of transportation capacity, and decomposed into various indicators to be calculated that can represent the quality of the driver's work, such as the time of dispatching, service points, and the time of listening to orders at a fixed price. Below, the time of dispatching is used as an example to illustrate. It can also be layered by driver dimension and time period dimension (layered dimension) to conduct data exploration. Through the comparison of historical data and the prediction of individual driver data, it can be concluded which driver layer groups or which individual driver's behavioral problems caused the data fluctuation of the analysis path, and then through the hit comparison with the attribution module, a reliable analysis conclusion of the cause of the fluctuation can be obtained.

[0066] In some embodiments, the stratification dimension includes at least one of a driver stratification dimension, a time period stratification dimension, and an order type stratification dimension. Data stratification is a way of decomposing data, which can more accurately analyze and locate data problems.

[0067] Exemplarily, the hierarchical decomposition of the driver stratification dimension is used as an example to illustrate that the hierarchical decomposition according to the driver stratification dimension can be mainly carried out by dividing the entire driver group into groups based on regular driver behavior. For example, according to the daily number of orders completed by the drivers, they can be divided into core drivers (daily number of orders completed is greater than 30), head drivers (daily number of orders completed is between 20-30), mid-level drivers (daily number of orders completed is between 10-20), and tail drivers (daily number of orders completed is less than 10); through this hierarchical decomposition of data from the driver stratification dimension, it can be seen which driver stratification groups' data performance under some analysis paths has caused the greatest data fluctuations, so as to locate which driver stratification group has caused the impact on the data fluctuations in the analysis path where it is located, and then it can be mutually corroborated with the fluctuation cause judgment of the attribution module, so as to draw the final fluctuation cause analysis conclusion.

[0068] For example, by stratifying according to the time period dimension, we can use historical data experience to determine which time period is the peak demand period within the 24-hour period of a day, which can be used as the peak time period stratification. We can also break it down into morning peak, noon peak, night peak, and off-peak stratification.

[0069] In some embodiments, fluctuation attribution analysis is performed based on the results of each path analysis to obtain fluctuation cause analysis conclusions for core operating indicators, including: determining fluctuation influencing factors, which include internal and external factors; collecting historical internal factor actual data corresponding to each internal factor and historical external factor actual data corresponding to each external factor; predicting the fluctuation trend of core operating indicators under different fluctuation influencing factors based on the historical internal factor actual data and the historical external factor actual data; comparing the path analysis results with the fluctuation trends under each fluctuation influencing factor to determine fluctuation cause analysis conclusions for core operating indicators.

[0070] In this embodiment, internal factors may include but are not limited to platform passenger-side prices, platform driver-side prices, etc., and external factors may include but are not limited to competitor passenger-side prices, competitor driver-side prices, weather effects, holiday effects, event effects, etc.

[0071] By collecting the actual data of historical internal factors corresponding to each internal factor and the actual data of historical external factors corresponding to each external factor, the fluctuation trend of core operating indicators under different fluctuation influencing factors is predicted based on the actual data of historical internal factors and the actual data of historical external factors. That is, by acquiring and processing the external and internal factor data, the actual situation is transformed into data performance, and the data performance of the actual situation (attribution hit module) and the path analysis results after data decomposition obtained by the analysis path are compared to obtain accurate cause hits.

[0072] For example, weather attribution analysis can recover past historical data, conduct data training through historical actual data performance and historical actual weather conditions, and obtain a calculation model for weather attribution analysis. Based on this calculation model, the data fluctuation trend of core operating indicators during various weather changes is obtained. Then, the abnormal trend can be further combined with human intervention to finally obtain the impact of weather as a fluctuation factor on the fluctuation trend of core operating indicators.

[0073] The standard for whether the impact of the volatility factor on the volatility trend of the core operating indicators is consistent with the path analysis results: the data after the time period stratification dimension or the driver stratification dimension and the objective performance of the volatility factor on the data need to be consistent. For example, the number of completed orders on the 8th increased, that is, the data had an upward fluctuation trend. The stratified group in this period had a greater impact on the fluctuation of the analysis path in which it was located. It also rained on the 8th. According to the historical external data of the weather volatility factor, it can be known that it rained from 9 to 10 o'clock on the 8th, and the volatility trend of the number of completed orders under the influence of the weather volatility factor between 9 and 10 o'clock did not increase. Therefore, this kind of fine-layer data decomposition did not hit the cause, that is, the conclusion of the fluctuation cause analysis is that it cannot be attributed to the rainy weather. By analyzing the data corresponding to different volatility factors, we can finally get the causes of the fluctuation of the core operating indicators, which volatility factors are associated with them, and then get the conclusion of the fluctuation cause analysis.

[0074] Below, reference Figure 3 As shown, Figure 3 The interactive sequence diagram of the method for analyzing fluctuations in online car-hailing operation data in some embodiments is shown below. Specifically, the following steps may be included:

[0075] S1: Update of basic operation data.

[0076] Through the data collection module, development interfaces and crawler technology are used to stably and efficiently collect data from various channels to the data warehouse and update the data regularly. The collected data includes but is not limited to API interface docking, log file analysis, web page crawling and other means. The collected data may include basic operating data closely related to the online car-hailing industry, such as driver and passenger data, holidays, weather, etc.

[0077] S2: Request basic operational data.

[0078] The algorithm module requests basic operational data from the data warehouse periodically, in real time, or in response to instructions.

[0079] S2.1: Return basic operational data.

[0080] In response to the request of the algorithm module, the data warehouse returns the basic operational data to the algorithm module.

[0081] S3: Trigger attribution analysis instructions.

[0082] The user of the client triggers the attribution analysis instruction.

[0083] S3.1: Request attribution analysis results.

[0084] The client requests the attribution analysis results from the server.

[0085] S3.2: Request the basic application data relied upon by the attribution analysis and the conclusions of the fluctuation cause analysis.

[0086] The server requests the algorithm module to provide the basic application data that the attribution analysis relies on and the analysis conclusions of the causes of fluctuations.

[0087] S3.3: Analysis path decomposition and attribution analysis.

[0088] The algorithm module splits the basic operating data according to different analysis paths to generate multiple groups of operating data to be calculated, inputs each group of operating data to be calculated into the corresponding fluctuation analysis model, analyzes and calculates each group of operating data to be calculated through each fluctuation analysis model, and obtains the path analysis results output by each fluctuation analysis model. According to the analysis results of each path, fluctuation attribution analysis is performed to obtain the fluctuation cause analysis conclusions for the core operating indicators.

[0089] S3.4: Analysis path decomposition and attribution analysis.

[0090] The algorithm module returns the path analysis results and fluctuation cause analysis conclusions to the server.

[0091] S3.5: Data assembly.

[0092] The server assembles the data to obtain data that can be adapted to the client.

[0093] S3.6: Return the attribution analysis results.

[0094] The server returns the assembled data as the attribution analysis result to the client that initiated the request.

[0095] S4: Customized hierarchical dimension and analysis path splitting dimension.

[0096] The client supports users to customize the hierarchical dimensions, specific division conditions of each hierarchical dimension, analysis paths, and analysis directions of each analysis path, etc.

[0097] The above application examples have at least the following characteristics:

[0098] 1. Multi-source data integration and processing: The system can access and integrate passenger flow data from different channels, as well as supply-side information such as vehicle and driver status from different transport suppliers, to achieve unified data management and cleaning, laying the foundation for subsequent analysis.

[0099] 2. Intelligent core indicator analysis: Through advanced data analysis algorithms, the system can automatically monitor and analyze key business indicators - such as the number of completed orders, etc. It can provide a real-time monitoring panel and conduct alarm analysis on historical abnormal fluctuations.

[0100] 3. Multi-dimensional cross-analysis: Machine learning technology can be used to analyze path decomposition and conduct detailed decomposition in the three directions of demand, supply, and efficiency. This can include driver behavior analysis, passenger behavior analysis, regional supply and demand matching evaluation, time period effect research and other multi-dimensional cross-analyses to identify abnormal fluctuation points and the contribution of each path, providing a basis for conclusions and cause verification.

[0101] 4. Causal reasoning and decision support: Combined with the causal inference model, the system not only displays the data results, but also conducts in-depth analysis of the reasons behind the data, such as whether the decrease in the number of completed orders in a specific time period is related to weather, event promotions, or competitor strategies, and provides the operation team with executable improvement suggestions and strategy adjustment plans.

[0102] 5. Dynamic strategy generation and optimization: Based on the analysis results, the system can dynamically generate operational strategy recommendations, such as dynamic pricing, driver incentives, passenger discount strategies, etc., and continuously optimize the strategy effects through a continuous feedback mechanism to achieve refined operations.

[0103] It should be understood that although Figure 2 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 2 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0104] In some embodiments, Figure 4 As shown, a device for analyzing fluctuations in online car-hailing operation data is provided, including: a data acquisition module 410, a data splitting module 420, a splitting calculation module 430 and an attribution hit module 440, wherein:

[0105] The data acquisition module 410 is used to respond to the attribution analysis instruction and acquire the basic operation data of the online car-hailing vehicle generated during the time period when the core operation indicators fluctuate abnormally;

[0106] The data splitting module 420 is used to split the basic operation data according to different analysis paths to generate multiple groups of operation data to be calculated, and input each group of operation data to be calculated into a corresponding fluctuation analysis model, each fluctuation analysis model corresponds to each analysis path;

[0107] The split calculation module 430 is used to analyze and calculate each group of operation data to be calculated by each fluctuation analysis model, and obtain the path analysis results output by each fluctuation analysis model;

[0108] The attribution hit module 440 is used to perform fluctuation attribution analysis based on the analysis results of each path, and obtain the fluctuation cause analysis conclusion for the core operating indicators.

[0109] In some embodiments, the data splitting module 420 determines the indicators to be calculated for each analysis path; and splits the basic operation data according to each indicator to be calculated to generate multiple groups of operation data to be calculated that match the calculation requirements of each indicator to be calculated.

[0110] In some embodiments, the split calculation module 430 decomposes the operation data to be calculated into layers according to different hierarchical dimensions; obtains historical operation data that is hierarchically divided into layers according to each hierarchical dimension; calculates the data fluctuation changes of each layer of the operation data to be calculated based on the historical operation data; and determines the hierarchical group whose fluctuation impact on the analysis path is greater than a preset threshold based on the data fluctuation changes of each layer.

[0111] In some embodiments, the attribution hit module 440 determines the fluctuation influencing factors, which include internal factors and external factors; collects historical internal factor real-time data corresponding to each internal factor and historical external factor real-time data corresponding to each external factor; predicts the fluctuation trend of core operating indicators under different fluctuation influencing factors based on the historical internal factor real-time data and the historical external factor real-time data; compares the path analysis results with the fluctuation trends under each fluctuation influencing factor to determine the fluctuation cause analysis conclusion for the core operating indicators.

[0112] For the specific limitations of the analysis device for fluctuations in online ride-hailing operation data, please refer to the limitations of the analysis method for fluctuations in online ride-hailing operation data above, which will not be repeated here. Each module in the above-mentioned analysis device for fluctuations in online ride-hailing operation data can be implemented in whole or in part through software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0113] In some embodiments, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 5 As shown. The computer device includes a processor, a memory, and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for analyzing fluctuations in online car-hailing operation data is implemented.

[0114] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0115] In some embodiments, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the following steps are implemented when the processor executes the computer program: in response to an attribution analysis instruction, basic operation data of online ride-hailing vehicles generated during a time period when abnormal fluctuations occur in core operation indicators; the basic operation data are split according to different analysis paths to generate multiple groups of operation data to be calculated, and each group of operation data to be calculated is input into a corresponding fluctuation analysis model, wherein each fluctuation analysis model corresponds to each analysis path; each group of operation data to be calculated is analyzed and calculated by each fluctuation analysis model to obtain path analysis results output by each fluctuation analysis model; fluctuation attribution analysis is performed based on the analysis results of each path to obtain fluctuation cause analysis conclusions for core operation indicators.

[0116] In some embodiments, when the processor executes the computer program, the following steps are also implemented: determining the indicators to be calculated for each analysis path; splitting the basic operation data according to each indicator to be calculated to generate multiple groups of operation data to be calculated that match the calculation requirements of each indicator to be calculated.

[0117] In some embodiments, when the processor executes the computer program, the following steps are also implemented: decomposing the operation data to be calculated into layers according to different layering dimensions; obtaining historical operation data that is layered according to each layering dimension; calculating the data fluctuation changes of each layer of the operation data to be calculated based on the historical operation data; and determining the layered group whose fluctuation impact on the analysis path is greater than a preset threshold based on the data fluctuation changes of each layer.

[0118] In some embodiments, when the processor executes the computer program, the following steps are also implemented: determining fluctuation influencing factors, which include internal factors and external factors; collecting historical internal factor actual data corresponding to each internal factor and historical external factor actual data corresponding to each external factor; predicting the fluctuation trend of core operating indicators under different fluctuation influencing factors based on the historical internal factor actual data and the historical external factor actual data; comparing the path analysis results with the fluctuation trends under each fluctuation influencing factor to determine the fluctuation cause analysis conclusion for the core operating indicators.

[0119] In some embodiments, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented: in response to an attribution analysis instruction, basic operating data of the online ride-hailing vehicle generated during a time period when abnormal fluctuations occur in core operating indicators; the basic operating data is split according to different analysis paths to generate multiple groups of operating data to be calculated, and each group of operating data to be calculated is input into a corresponding fluctuation analysis model, and each fluctuation analysis model corresponds to each analysis path; each group of operating data to be calculated is analyzed and calculated by each fluctuation analysis model to obtain path analysis results output by each fluctuation analysis model; fluctuation attribution analysis is performed according to the analysis results of each path to obtain a fluctuation cause analysis conclusion for the core operating indicator.

[0120] In some embodiments, when the computer program is executed by the processor, the following steps are also implemented: determining the indicators to be calculated for each analysis path; splitting the basic operation data according to each indicator to be calculated to generate multiple groups of operation data to be calculated that match the calculation requirements of each indicator to be calculated.

[0121] In some embodiments, when the computer program is executed by the processor, the following steps are also implemented: decomposing the operation data to be calculated into layers according to different hierarchical dimensions; obtaining historical operation data that is hierarchically divided according to each hierarchical dimension; calculating the data fluctuation changes of each layer of the operation data to be calculated based on the historical operation data; and determining the hierarchical groups whose fluctuation impact on the analysis path is greater than a preset threshold based on the data fluctuation changes of each layer.

[0122] In some embodiments, when the computer program is executed by the processor, the following steps are also implemented: determining fluctuation influencing factors, which include internal factors and external factors; collecting historical internal factor actual data corresponding to each internal factor and historical external factor actual data corresponding to each external factor; predicting the fluctuation trend of core operating indicators under different fluctuation influencing factors based on the historical internal factor actual data and the historical external factor actual data; comparing the path analysis results with the fluctuation trends under each fluctuation influencing factor to determine the fluctuation cause analysis conclusion for the core operating indicators.

[0123] Those skilled in the art can understand that all or part of the processes in the methods for implementing the above embodiments can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0124] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0125] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

[0126] It should be noted that in the embodiments of the present application, data related to user information or user data (for example, basic operating data, etc.) must be obtained and processed after the user's authorization and consent. When the embodiments of the present application are applied to specific products or technologies, the user's permission or consent must be obtained, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

Claims

1. A method for analyzing fluctuations in online car-hailing operation data, the method comprising: In response to the attribution analysis instruction, obtain the basic operation data of the online car-hailing service generated during the time period when the core operation indicators fluctuate abnormally; Splitting the basic operation data according to different analysis paths to generate multiple groups of operation data to be calculated, inputting each group of the operation data to be calculated into a corresponding fluctuation analysis model, each fluctuation analysis model corresponds to each analysis path; By using each of the fluctuation analysis models, each group of the to-be-calculated operation data is analyzed and calculated respectively, so as to obtain the path analysis results outputted by each of the fluctuation analysis models respectively; A fluctuation attribution analysis is performed based on the results of each of the path analyses to obtain a fluctuation cause analysis conclusion for the core operating indicators.

2. The method according to claim 1, characterized in that: The step of splitting the basic operation data according to different analysis paths to generate multiple groups of operation data to be calculated includes: Determining the indicators to be calculated for each of the analysis paths; The basic operation data is split according to each of the indicators to be calculated to generate multiple groups of operation data to be calculated that match the calculation requirements of each of the indicators to be calculated.

3. The method according to claim 2, characterized in that The analysis path includes a demand path on the passenger side, a supply path on the driver side, and an efficiency evaluation path.

4. The method according to claim 3, characterized in that The indicators to be calculated for the demand path on the passenger side include at least one of the order quantity, matching quantity, response quantity, order completion quantity, cancellation quantity and response rate; the supply path on the driver side includes at least one of the number of drivers dispatched, the number of drivers completing orders and the order completion rate; the efficiency evaluation path includes at least one indicator characterizing the driver's work quality.

5. The method according to claim 1, characterized in that The analyzing and calculating each group of the to-be-calculated operation data by each of the fluctuation analysis models to obtain the path analysis results output by each of the fluctuation analysis models respectively includes: Decomposing the to-be-calculated operational data in different layers according to different layer dimensions; Acquire historical operation data that is stratified according to each of the stratification dimensions; Calculate the data fluctuation change of each layer of the to-be-calculated operating data according to the historical operating data; According to the data fluctuation change of each layer, a layer group whose fluctuation influence on the analysis path is greater than a preset threshold is determined.

6. The method according to claim 1, characterized in that The hierarchical dimensions include at least one of a driver hierarchical dimension, a time period hierarchical dimension, and an order type hierarchical dimension.

7. The method according to any one of claims 1 to 6, characterized in that: The fluctuation attribution analysis is performed according to the path analysis results to obtain the fluctuation cause analysis conclusion for the core operating indicators, including: Determining fluctuation influencing factors, wherein the fluctuation influencing factors include internal factors and external factors; Collecting the historical internal factor real-time data corresponding to each of the internal factors and the historical external factor real-time data corresponding to each of the external factors; Predicting the fluctuation trend of the core operating indicators under different fluctuation influencing factors according to the historical internal cause real-time data and the historical external cause real-time data; The path analysis results are compared with the fluctuation trends under each of the fluctuation influencing factors to determine the fluctuation cause analysis conclusions for the core operating indicators.

8. An analysis device for fluctuations in online car-hailing operation data, characterized in that: The device comprises: A data acquisition module, for responding to the attribution analysis instruction and acquiring the basic operation data of the online car-hailing service generated during the period when the core operation indicators fluctuate abnormally; A data splitting module, used for splitting the basic operation data according to different analysis paths to generate multiple groups of operation data to be calculated, and inputting each group of the operation data to be calculated into a corresponding fluctuation analysis model, each fluctuation analysis model corresponds to each analysis path; A split calculation module, used to analyze and calculate each group of the to-be-calculated operation data respectively through each of the fluctuation analysis models, and obtain the path analysis results output by each of the fluctuation analysis models respectively; The attribution hit module is used to perform fluctuation attribution analysis based on the results of each path analysis to obtain fluctuation cause analysis conclusions for the core operating indicators.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images