Order analysis-based online car-hailing management method and system

Abstract

The application discloses a car-hailing management method and system based on order analysis, relates to the technical field of car-hailing management, and acquires order data and a plurality of data related to car-hailing operation from a car-hailing management platform at regular time intervals, generates a change index for each partition after analyzing the partitioned order data and the plurality of data through a correlation model, generates an updated partition list after updating and sorting all the partitions in descending order of the change index, and judges whether the partition has a change anomaly through a comparison result of the change index and a change threshold value, and generates a management suggestion for the partition when it is judged that the partition has a change anomaly. The management system can analyze whether the car-hailing operation has an anomaly at regular time intervals, so that the car-hailing management platform can be warned and reminded in advance when the car-hailing operation has an anomaly, the management effect and management efficiency of the car-hailing are effectively improved, and user or driver loss is avoided.

Description

Technical Field

[0001] This invention relates to the field of ride-hailing management technology, specifically to a ride-hailing management method and system based on order analysis. Background Technology

[0002] Ride-hailing management refers to a series of measures and regulations for supervising and managing the ride-hailing industry. With the development of Internet technology, the ride-hailing industry has developed rapidly worldwide. This industry connects passengers and drivers through online platforms, providing convenient and flexible transportation services. However, due to the special nature of the ride-hailing industry, certain management and supervision are needed to ensure public safety, maintain market order, and protect consumer rights.

[0003] The existing technology has the following drawbacks:

[0004] To avoid over-managing ride-hailing services and increasing management costs, existing management systems typically manage ride-hailing services at relatively long intervals (e.g., once a month or once a quarter). However, due to the real-time and uncertain nature of ride-hailing operations, if a periodic management approach is adopted, the existing management system will not issue warnings or alerts when abnormalities occur. Continuing with periodic management will reduce the effectiveness and efficiency of the management system for ride-hailing services, leading to the loss of users or drivers.

[0005] Based on this, the present invention proposes a ride-hailing management method and system based on order analysis, which can perform management analysis on ride-hailing vehicles at regular intervals to determine whether it is necessary to issue warnings and reminders to the administrator, so that management can be carried out in advance when there are abnormalities in the operation of ride-hailing vehicles, thereby effectively improving the management effect and efficiency of ride-hailing vehicles. Summary of the Invention

[0006] The purpose of this invention is to provide a ride-hailing management method and system based on order analysis to address the shortcomings in the prior art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a ride-hailing management method based on order analysis, the management method comprising the following steps:

[0008] The management system obtains the number of districts in the current city through the city management platform, generates a unique identifier for each district, performs an initial sorting of each district based on its economic status, and generates an initial district list.

[0009] Regularly retrieve order data and various data related to ride-hailing operations from the ride-hailing management platform. After analyzing the order data and various data of each partition through a correlation model, generate a change index for each partition.

[0010] After updating and sorting all partitions according to their change index from largest to smallest, an updated partition list is generated. Then, the partitions are judged to have any abnormal changes by comparing the change index with the change threshold.

[0011] When a partition is found to have abnormal changes, the management system generates management suggestions for that partition and sends the updated sorting list and management suggestion information to the ride-hailing management platform.

[0012] When conducting regular overall management of ride-hailing services, the historical change index of each zone is obtained. The historical change index is analyzed to generate a management value for each zone. Each zone is sorted from largest to smallest according to the management value to generate a management zone list. The management system generates management policies based on the management zone list and sends them to the ride-hailing management platform.

[0013] In a preferred embodiment, the management system periodically obtains order data and various data related to ride-hailing operations from the ride-hailing management platform. The order data includes the order acceptance change coefficient, and the various data include the driver account cancellation rate, the duration deviation index, and the order completion rate.

[0014] In a preferred embodiment, the establishment of the association model includes the following steps:

[0015] The order acceptance variation coefficient, driver account cancellation rate, time deviation index, and order completion rate are normalized to generate the variation index. The formula for calculating the variation index is as follows: In the formula, bd is the degree of change index, JDB is the order change coefficient, ZHX is the driver account cancellation rate, SCP is the duration deviation index, WDL is the order completion rate, and α, β, γ, and τ are the proportional coefficients of the order change coefficient, driver account cancellation rate, duration deviation index, and order completion rate, respectively. α, β, γ, and τ are all greater than 0. After calculating the degree of change index, the correlation model is established.

[0016] In a preferred embodiment, generating a variability index for each partition includes the following steps:

[0017] The management system will input the order change coefficient, machine account cancellation rate, duration deviation index and order completion rate obtained periodically into the correlation model;

[0018] The correlation model, after comprehensively analyzing the order change coefficient, machine account cancellation rate, time deviation index, and order completion rate, outputs a change index for each partition.

[0019] In a preferred embodiment, determining whether a partition exhibits abnormal changes by comparing the change index with the change threshold includes the following steps:

[0020] If the degree of change index is less than or equal to the degree of change threshold, it is determined that there is no abnormal change in the partition.

[0021] If the change index is greater than the change threshold, it is determined that there is an abnormal change in the partition, and the management system sends a warning signal to the ride-hailing management platform.

[0022] In a preferred embodiment, when periodically managing ride-hailing services as a whole, the historical change index of each zone is obtained, the historical change index is analyzed to generate a management value for each zone, and each zone is sorted from largest to smallest according to the management value to generate a management zone list, including the following steps:

[0023] The management assignment for each zone is calculated by combining the standard deviation of the variability index, the average variability index, and the variability index of the management system for the previous time period, which is used for overall management of ride-hailing services. The expression is as follows:

[0024]

[0025] In the formula, GLZ represents the management assignment, and bd... jq To manage ride-hailing services as a whole, the change index of the previous time period is used. avg BQ is the average of the degree of change index. c bd represents the standard deviation of the variability index. 阈值 BQ is the threshold for variability. 阈值 The standard deviation threshold;

[0026] The larger the management assignment GLZ of a partition, the more necessary it is to manage that partition first. After obtaining the management assignment GLZ of all partitions, sort all partitions in descending order of their management assignment GLZ values ​​to generate a list of managed partitions.

[0027] In a preferred embodiment, the logic for obtaining the average value and standard deviation of the variability index is as follows:

[0028] The time interval at which the management system periodically retrieves order data and various data related to ride-hailing operations from the ride-hailing management platform is marked as Td;

[0029] The time interval for the management system to periodically conduct overall management of ride-hailing vehicles is marked as Tq;

[0030] Then the number of historical change indexes obtained for the partition is n = Tq / Td-1:

[0031] The expressions for calculating the mean and standard deviation of the variability index are as follows:

[0032]

[0033] In the formula, i = {1, 2, 3, ..., n}, n represents the number of historical change indices for the partition, and n is a positive integer. i bd represents the i-th degree of change index in the set of degree of change indices. avg BQ represents the average value of the degree of change index. c It represents the standard deviation of the variability index.

[0034] In a preferred embodiment, generating an updated partition list after updating and sorting all partitions according to their degree of change index from largest to smallest includes the following steps:

[0035] The obtained partition information is sorted using a sorting algorithm. The partition information is sorted from largest to smallest according to the degree of change through merge sort.

[0036] The sorted partition information is organized into an updated partition list, which contains information for each partition, including partition name, change index, and partition number.

[0037] The updated list of partitions can be stored in a database or displayed directly.

[0038] In a preferred embodiment, the calculation logic of the order change coefficient is as follows: the period when the number of user orders minus the number of ride-hailing vehicles exceeds the difference threshold is used as the period for order oversupply warning. During the monitoring period, the longer the period for order oversupply warning is, the more likely it is to lead to user churn.

[0039] The period during which the number of ride-hailing vehicles minus the number of user orders exceeds a threshold is used as the period for order overrun warning. Within the monitoring period, the longer the period for order overrun warning is, the more likely it is to lead to the loss of ride-hailing drivers.

[0040] The order acceptance change coefficient is obtained by integrating the periods of order oversupply warning and order undersupply warning. The calculation expression is as follows: In the formula, D(t) represents the real-time change in the partitioned orders, [t x , t y [t] is the period for issuing order overage warnings. i , t j This is the period for issuing early warnings of excess orders.

[0041] This invention also provides a ride-hailing management system based on order analysis, including a partition identification module, an initial sorting module, a data acquisition module, a data analysis module, an update sorting module, a judgment module, and a management sorting module.

[0042] Zone Identification Module: Obtain the number of zones in the current city through the city management platform, and generate a unique identifier for each zone. Each identifier includes the zone name, zone economic status, and zone number.

[0043] Initial sorting module: Performs an initial sorting of each partition based on economic conditions, generating an initial partition list;

[0044] Data acquisition module: Periodically retrieves order data and various data related to ride-hailing operations from the ride-hailing management platform;

[0045] Data Analysis Module: After analyzing order data and multiple data items in each partition using a correlation model, it generates a change index for each partition.

[0046] Update sorting module: After updating and sorting all partitions according to the degree of change index from large to small, an updated partition list is generated;

[0047] Judgment Module: The system determines whether there are any abnormal changes in a partition by comparing the change index with the change threshold. When an abnormal change is found in a partition, the system generates management suggestions for that partition and sends the updated sorting list and management suggestion information to the ride-hailing management platform.

[0048] Management and sorting module: When performing overall management of ride-hailing services periodically, the module obtains the historical change index of each partition, analyzes the historical change index to generate a management value for each partition, sorts each partition from largest to smallest according to the management value, generates a management partition list, generates management strategies based on the management partition list, and sends them to the ride-hailing management platform.

[0049] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0050] 1. This invention involves a management system that periodically retrieves order data and various data related to ride-hailing operations from a ride-hailing management platform. After analyzing the order data and other data for each region using a correlation model, a change index is generated for each region. All regions are first updated and sorted according to their change index from highest to lowest, generating an updated region list. Then, the change index is compared with a change threshold to determine if any region exhibits abnormal changes. When an abnormality is detected, the management system generates management suggestions for that region and sends the updated sorting list and management suggestions to the ride-hailing management platform. This management system can periodically analyze ride-hailing operations for anomalies, thus providing early warnings to the ride-hailing management platform when anomalies are detected, effectively improving the management effectiveness and efficiency of ride-hailing services and preventing user or driver churn.

[0051] 2. When periodically managing ride-hailing services, this invention obtains the historical change index of each zone, analyzes the historical change index to generate a management value for each zone, sorts each zone from largest to smallest according to the management value, generates a management zone list, and the management system generates management strategies based on the management zone list and sends them to the ride-hailing management platform. This enables the ride-hailing management platform to prioritize the management of zones with poor ride-hailing service performance, further improving the efficiency of ride-hailing service management. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0053] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0055] Example 1: Please refer to Figure 1 As shown in this embodiment, the ride-hailing management method based on order analysis includes the following steps:

[0056] The city management platform is used to obtain the number of districts in the current city and generate a unique identifier for each district. Each identifier includes the district name, the district's economic status, and the district number. This process includes the following steps:

[0057] Understand the city management platform's API documentation: Obtain the city management platform's API documentation, which includes detailed descriptions of endpoints, authentication methods, parameters, and response formats. Review the documentation to understand how to obtain the city's zoning information.

[0058] To obtain the number of zones: Use the corresponding API endpoint provided by the platform (which may be a path like / city / {city_id} / zones) and authorized authentication information (such as an API key or OAuth token) to initiate an HTTP request to obtain the number of zones for the current city. Parse the API response to obtain the number of zones.

[0059] Generate a unique identifier for each partition: For each partition, generate a unique identifier using a unique identifier algorithm. For example, use a UUID library.

[0060] Retrieve partition information: Iterate through each partition and initiate an HTTP request using the platform's provided API endpoint (e.g., / city / {city_id} / zones / {zone_id}) to obtain detailed information for each partition. The request should include the partition's unique identifier and authorization information.

[0061] Generate Identifier: Integrate the detailed information of each partition into the previously generated unique identifier. Create a data structure or object to store information such as partition name, partition economic status, and partition number.

[0062] The initial sorting of each partition based on economic conditions generates an initial partition list, including the following steps:

[0063] Obtaining zone information: Using the obtained city management platform API, obtain detailed information for each zone, including zone name, zone economic status, and zone number.

[0064] Parse partition information: Analyze the economic data of each partition to accurately extract and understand the economic situation of each partition.

[0065] Sorting based on economic status: Initially sort all partitions using their economic status data. Then, use merge sort to perform the sorting when retrieving partition information.

[0066] Implement the merge sort algorithm, a divide-and-conquer sorting algorithm that breaks down a problem into smaller subproblems and then merges the solutions to the original problem. In this scenario, economic data is used as the sorting basis. The merge sort algorithm is used to sort the obtained partition information. During the sorting process, the partition information is sorted in ascending or descending order according to economic status.

[0067] Generate an initial partition list: Organize the sorted partition information into an initial partition list. This list can be a data structure containing information for each partition, including partition name, partition economic status, partition number, etc.

[0068] Store or display the initial partition list: Store the initial partition list in a database, a file, or display it directly on the user interface for later use or analysis.

[0069] The management system periodically retrieves order data and various data related to ride-hailing operations from the ride-hailing management platform. After analyzing the order data and various data points by partition using a correlation model, it generates a change index for each partition. All partitions are then updated and sorted according to their change index from largest to smallest, generating an updated partition list. This process includes the following steps:

[0070] Sorting Algorithm: Implement a sorting algorithm. Use merge sort to sort partition information in descending order of degree of change.

[0071] Sort the partitions: Use a sorting algorithm to sort the obtained partition information to ensure that they are arranged in descending order of the degree of change index.

[0072] Generate an updated partition list: Organize the sorted partition information into an updated partition list. This list contains information for each partition, including partition name, variability index, partition number, etc.

[0073] Store or display the updated partition list: Store the updated partition list in a database, a file, or display it directly on the user interface for later use or analysis.

[0074] The system determines whether a partition exhibits abnormal changes by comparing the change index with the change threshold. When an abnormal change is detected, the system generates management suggestions for that partition and sends the updated sorting list and management suggestions to the ride-hailing management platform, including the following steps:

[0075] Generate management suggestions: When the system detects abnormal changes, it generates corresponding management suggestions. For example, it may suggest that ride-hailing services in that area need to be managed in advance, which may include suggestions such as increasing the number of vehicles or optimizing the scheduling algorithm.

[0076] Integrate management suggestion information: Combine the generated management suggestion information with the updated sorted list to form a data structure containing management suggestions. Specifically, this is a JSON object or other data format, including information such as partition name, change index, partition number, and management suggestion.

[0077] Sending to the ride-hailing management platform: Utilize the API provided by the ride-hailing management platform to send the integrated information to the platform using an HTTP POST request or other communication protocols. Authentication information provided by the platform is also required.

[0078] This application utilizes a management system to periodically retrieve order data and various data related to ride-hailing operations from a ride-hailing management platform. After analyzing the order data and other data for each region using a correlation model, a change index is generated for each region. All regions are first updated and sorted according to their change index from highest to lowest, generating an updated region list. Then, the change index is compared with a change threshold to determine if any region exhibits abnormal changes. When an abnormality is detected, the management system generates management suggestions for that region and sends the updated sorting list and management suggestions to the ride-hailing management platform. This management system can periodically analyze ride-hailing operations for anomalies, thus providing early warnings to the ride-hailing management platform when anomalies are detected, effectively improving the management effectiveness and efficiency of ride-hailing services and preventing user or driver churn.

[0079] This application, when periodically managing ride-hailing services as a whole, obtains the historical change index of each zone, analyzes the historical change index to generate a management value for each zone, sorts each zone from largest to smallest according to the management value, generates a management zone list, and the management system generates management policies based on the management zone list and sends them to the ride-hailing management platform. This enables the ride-hailing management platform to prioritize the management of zones with poor ride-hailing service performance, further improving the efficiency of ride-hailing service management.

[0080] Example 2: The management system periodically obtains order data and various data related to ride-hailing operations from the ride-hailing management platform. After analyzing the order data and various data of each partition through the association model, a change index is generated for each partition.

[0081] The management system periodically obtains order data and various data related to ride-hailing operations from the ride-hailing management platform. The order data includes the order acceptance change coefficient, and the various data includes driver account cancellation rate, time deviation index, and order completion rate.

[0082] After analyzing the periodically acquired order change coefficient, machine account cancellation rate, duration deviation index, and order completion rate, the output is a change index for each partition.

[0083] The establishment of the association model includes the following steps:

[0084] The order acceptance variation coefficient, driver account cancellation rate, time deviation index, and order completion rate are normalized to generate the variation index. The formula for calculating the variation index is as follows: In the formula, bd is the degree of change index, JDB is the order change coefficient, ZHX is the driver account cancellation rate, SCP is the time deviation index, WDL is the order completion rate, and α, β, γ, and τ are the proportional coefficients of the order change coefficient, driver account cancellation rate, time deviation index, and order completion rate, respectively, and α, β, γ, and τ are all greater than 0.

[0085] After calculating the degree of change index, the correlation model is established.

[0086] The calculation logic for the order change coefficient is as follows: During the monitoring period, when the number of user orders minus the number of ride-hailing vehicles in a region exceeds the difference threshold, it will lead to an extension of the user's order acceptance time or no ride-hailing vehicles accepting the order. When these problems occur, they will lead to user churn on the ride-hailing management platform. The period when the number of user orders minus the number of ride-hailing vehicles exceeds the difference threshold is designated as the period of order oversupply warning. During the monitoring period, the longer the period of order oversupply warning, the more likely it is to lead to user churn.

[0087] During the monitoring period, when the number of ride-hailing vehicles in a region minus the number of user orders exceeds the difference threshold, ride-hailing drivers will not receive orders for a long time, which will cause them to lose confidence in the ride-hailing management platform and may lead to a decrease in the number of ride-hailing drivers. Therefore, the period when the number of ride-hailing vehicles minus the number of user orders exceeds the difference threshold is designated as the period for order shortage warning. During the monitoring period, the longer the period of order shortage warning, the more likely it is to cause ride-hailing driver churn.

[0088] The order acceptance change coefficient is obtained by integrating the periods of order oversupply warning and order undersupply warning. The calculation expression is as follows: In the formula, D(t) represents the real-time change in the partitioned orders, [t x , t y [t] is the period for issuing order overage warnings. i , t j This is the period for issuing early warnings of excess orders.

[0089] The formula for calculating the driver account cancellation rate is: ZHX = SX / Δt, where SX represents the number of driver account cancellations received by the ride-hailing management platform during the time period Δt, and Δt is the monitoring time period. The higher the driver account cancellation rate, the more ride-hailing drivers cancel their accounts during the monitoring time period. Specifically, this is reflected in the following:

[0090] Driver satisfaction issues: The increased driver account cancellation rate may be due to driver dissatisfaction with the service platform, including dissatisfaction with the dispatch algorithm, fee structure, and support services. Low satisfaction may lead drivers to choose to cancel their accounts.

[0091] Income issues: Drivers may choose to cancel their accounts due to income problems, which may be caused by factors such as price adjustments, intense competition, and a decrease in order volume.

[0092] Poor working conditions: Poor working conditions may include long waiting times for orders and difficulty in obtaining orders during peak hours. These conditions may lead to driver turnover.

[0093] The calculation logic of the Duration Deviation Index (SCP) is as follows: obtain the order acceptance delay time of ride-hailing vehicles within the monitoring period. The order acceptance delay time is obtained by subtracting the user's order placement time from the driver's order acceptance time. The order acceptance delay time of all ride-hailing vehicles belonging to the same ride-hailing management platform is added together to obtain the total order acceptance delay time.

[0094] During the monitoring period, the order completion delay time of ride-hailing vehicles is obtained. The order completion delay time is obtained by subtracting the estimated completion time from the actual completion time of the order. The total order completion delay time is obtained by adding the order completion delay times of all ride-hailing vehicles under the same ride-hailing management platform.

[0095] Finally, the total order acceptance delay time is added to the total order completion delay time to obtain the duration deviation index (SCP). The larger the duration deviation index, the slower the ride-hailing vehicle's order acceptance time or the slower the ride-hailing vehicle's order completion time is within the monitoring period. The resulting impacts include:

[0096] Decreased user experience: Users may have to wait longer to successfully hail a car or complete a trip, which may lead to a decline in user experience and affect user satisfaction with ride-hailing services.

[0097] Increased order churn: Users may cancel orders due to long wait times, especially in highly competitive markets. Increased order churn can reduce platform revenue.

[0098] Increased passenger complaints: Users may become dissatisfied due to slow service, increasing the likelihood of complaints. This could negatively impact the platform's brand reputation.

[0099] Decreased driver income: Increased waiting time for orders may reduce drivers' effective working hours, thus lowering their income levels. This can lead to driver dissatisfaction with the platform and increase the likelihood of drivers canceling their accounts.

[0100] Decreased operational efficiency: Extended order acceptance and completion times may indicate inefficiencies in scheduling and route planning, leading to a decline in overall operational efficiency.

[0101] Market share decline: Long waiting times and travel times may lead users to choose other modes of transportation, such as public transportation or driving, resulting in a decline in the market share of ride-hailing platforms.

[0102] Competitive disadvantage: If other competitors offer shorter pick-up and trip times, ride-hailing platforms may lose their competitive advantage in the market.

[0103] Regulatory compliance issues: Some regions may have regulations governing the service and response times of ride-hailing services. Long waiting times and trips may lead to platforms violating regulations and facing compliance issues.

[0104] Damaged brand image: Users' trust and recognition of ride-hailing platforms may be affected, and the brand image may be damaged, thereby affecting the platform's market position.

[0105] The calculation formula for the order completion rate is: WDL = DS / Δt, where DS represents the number of completed orders received by the ride-hailing management platform within the time period Δt, and Δt is the monitoring time period. The lower the order completion rate, the fewer orders the ride-hailing service completes within the monitoring time period. Specifically:

[0106] Service quality issues: A low completion rate may reflect poor service quality, leading to passenger cancellations or drivers ending services early. This could be due to driver lateness, poor vehicle condition, or poor service attitude.

[0107] Insufficient vehicles: If the number of vehicles is insufficient to meet user demand during the monitoring period, orders may not be completed in a timely manner. This may require adjusting the vehicle allocation strategy and increasing the number of vehicles deployed.

[0108] Traffic congestion or inefficient routes: High traffic congestion or drivers choosing inefficient routes can extend order completion times, thus reducing order completion rates. Optimizing scheduling algorithms is necessary to improve efficiency.

[0109] User experience issues: User experience issues may lead to order cancellations or passengers ending their trips early. These may include problems such as inconvenience in using the app and lack of price transparency.

[0110] Competitive pressure: In a fiercely competitive market, if competitors offer more attractive deals or more efficient services, the order completion rate may decrease.

[0111] Determining whether a partition exhibits abnormal changes by comparing the change index with the change threshold includes the following steps:

[0112] If the degree of change index is less than or equal to the degree of change threshold, it is determined that there is no abnormal change in the partition.

[0113] If the change index is greater than the change threshold, it is determined that there is an abnormal change in the partition, and the management system sends a warning signal to the ride-hailing management platform.

[0114] When conducting regular overall management of ride-hailing services, the historical change index of each zone is obtained. This historical change index is analyzed to generate a management value for each zone. The zones are then sorted from largest to smallest based on their management values ​​to generate a management zone list. The management system then generates management policies based on this list and sends them to the ride-hailing management platform. This process includes the following steps:

[0115] The time interval at which the management system periodically retrieves order data and various data related to ride-hailing operations from the ride-hailing management platform is marked as Td;

[0116] The time interval for the management system to periodically conduct overall management of ride-hailing vehicles is marked as Tq;

[0117] Then obtain the number of historical change indices for the partition n = Tq / Td-1; establish a change index set for all historical change indices of the partition, calculate the standard deviation and average of the historical change indices of the partition, and obtain the change index of the management system for the overall management of ride-hailing in the partition in the previous time period.

[0118] The management assignment for each zone is calculated by combining the standard deviation of the variability index, the average variability index, and the variability index of the management system for the previous time period, which is used for overall management of ride-hailing services. The expression is as follows:

[0119]

[0120] In the formula, GLZ represents the management assignment, and bd... jq To manage ride-hailing services as a whole, the change index of the previous time period is used. avg BQ is the average of the degree of change index. c bd represents the standard deviation of the variability index. 阈值 BQ is the threshold for variability. 阈值 The standard deviation threshold;

[0121] When bd avg ≤bd 阈值 ∧BQ c ≤BQ 阈值 When the average value of the change index of the partition is less than or equal to the change threshold, and the standard deviation of the change index is less than or equal to the standard deviation threshold, it indicates that there are no obvious abnormal changes in the operation of ride-hailing in the partition, and the fluctuation of the historical change index is small (i.e., the overall operation of ride-hailing in the partition is stable).

[0122] When bd avg ≤bd 阈值 ∧BQ c >BQ 阈值When the mean value of the change index of the partition is less than or equal to the change threshold, and the standard deviation of the change index is greater than the standard deviation threshold, it indicates that the change in the operation of ride-hailing services in the partition is abnormally small, but the historical change index fluctuates greatly (i.e., some change indices are greater than the change threshold).

[0123] When bd avg >bd 阈值 ∧BQ c >BQ 阈值 When the mean value of the change index of the partition is greater than the change threshold, and the standard deviation of the change index is greater than the standard deviation threshold, it indicates that the ride-hailing operation of the partition is abnormally large, but the historical change index fluctuates greatly (i.e., some change indices are less than or equal to the change threshold).

[0124] When bd avg ≤bd 阈值 ∧BQ c ≤BQ 阈值 When the average value of the variability index of a region is less than or equal to the variability threshold, and the standard deviation of the variability index is less than or equal to the standard deviation threshold, it indicates that there are obvious abnormal changes in the operation of ride-hailing services in the region, and the historical variability index fluctuates little (i.e., the overall operation of ride-hailing services in the region is unstable). The region needs to be prioritized for management.

[0125] The average value of the degree of change index bd avg and the standard deviation of the variability index BQ c The calculation expression is:

[0126]

[0127] In the formula, i = {1, 2, 3, ..., n}, n represents the number of historical change indices for the partition, and n is a positive integer. i bd represents the i-th degree of change index in the set of degree of change indices. avg This represents the average value of the degree of change index;

[0128] As can be seen from the calculation logic of the variability index bd, the variability index bd indicates that the more serious the abnormal changes in the operation of ride-hailing in a partition, the larger the management assignment GLZ of the partition, the more necessary it is to manage the partition first. After obtaining the management assignment GLZ of all partitions, all partitions are sorted from largest to smallest according to the management assignment GLZ to generate a list of managed partitions.

[0129] To better understand the above solution, we will provide the following detailed explanation:

[0130] Assume that the time interval between the management system's periodic retrieval of order data and various data related to ride-hailing operations from the ride-hailing management platform is marked as Td, and Td = 24h;

[0131] The time interval for the management system to conduct overall management of ride-hailing vehicles is marked as Tq, and Tq = 720h;

[0132] The number of historical change indices of the partition obtained is n = Tq / Td-1 = 720 / 24-1 = 29, that is, the number of historical change indices of the partition obtained is 29.

[0133] The management system manages ride-hailing services as a whole. The previous time period is the historical change index of the 29th partition.

[0134] The management system generates management policies based on the management partition list and sends them to the ride-hailing management platform, including the following steps:

[0135] The management system generates a management strategy based on the management partition list: the higher a partition appears in the management partition list, the more priority it needs to be managed.

[0136] The management strategy will then be sent to the ride-hailing management platform.

[0137] Example 3: The ride-hailing management system based on order analysis described in this example includes a partition identification module, an initial sorting module, a data collection module, a data analysis module, an update sorting module, a judgment module, and a management sorting module.

[0138] The zoning identification module obtains the number of zoning in the current city through the city management platform and generates a unique identifier for each zoning. Each identifier includes the zoning name, the economic status of the zoning, and the zoning number. The zoning identification information is sent to the initial sorting module, the update sorting module, and the management sorting module.

[0139] Initial sorting module: Performs an initial sorting of each partition based on economic conditions, generating an initial partition list;

[0140] Data acquisition module: Periodically retrieves order data and various data related to ride-hailing operations from the ride-hailing management platform, and sends the order data and other data to the data analysis module;

[0141] Data Analysis Module: After analyzing the order data and multiple data items of each partition through the association model, a change index is generated for each partition. The change index is then sent to the update sorting module, the management sorting module, and the judgment module.

[0142] Update sorting module: After updating and sorting all partitions according to the degree of change index from large to small, an updated partition list is generated;

[0143] Judgment Module: The system determines whether there are any abnormal changes in a partition by comparing the change index with the change threshold. When an abnormal change is found in a partition, the system generates management suggestions for that partition and sends the updated sorting list and management suggestion information to the ride-hailing management platform.

[0144] Management and sorting module: When performing overall management of ride-hailing services periodically, the module obtains the historical change index of each partition, analyzes the historical change index to generate a management value for each partition, sorts each partition from largest to smallest according to the management value, generates a management partition list, generates management strategies based on the management partition list, and sends them to the ride-hailing management platform.

[0145] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0146] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," 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, 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.

[0147] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for managing online car-hailing based on order analysis, characterized in that: The management method includes the following steps: The management system obtains the number of districts in the current city through the city management platform, generates a unique identifier for each district, performs an initial sorting of each district based on its economic status, and generates an initial district list. Regularly retrieve order data and various data related to ride-hailing operations from the ride-hailing management platform. After analyzing the order data and various data of each partition through a correlation model, generate a change index for each partition. After updating and sorting all partitions according to their change index from largest to smallest, an updated partition list is generated. Then, the partitions are judged to have any abnormal changes by comparing the change index with the change threshold. When a partition is found to have abnormal changes, the management system generates management suggestions for that partition and sends the updated sorting list and management suggestion information to the ride-hailing management platform. When regularly managing ride-hailing services as a whole, the historical change index of each zone is obtained, the historical change index is analyzed to generate a management value for each zone, and each zone is sorted from largest to smallest according to the management value to generate a management zone list. The management system generates management strategies based on the management zone list and sends them to the ride-hailing management platform. The establishment of the association model includes the following steps: The order receiving change coefficient, the driver account cancellation rate, the time length deviation index and the order completion rate are normalized to generate a change degree index, and the calculation expression of the change degree index is: ; in the formula, the change degree index is the order receiving change coefficient is the driver account cancellation rate is the time length deviation index is the order completion rate is , , , respectively the order receiving change coefficient, the driver account cancellation rate, the time length deviation index and the order completion rate are proportional coefficients, and , , , all greater than 0, the change degree index is calculated, and the correlation model is established; When conducting regular overall management of ride-hailing services, the historical change index of each zone is obtained. The historical change index is analyzed to generate a management value for each zone. Based on the management value, each zone is sorted from largest to smallest to generate a list of managed zones. This process includes the following steps: The management assignment for each zone is calculated by combining the standard deviation of the variability index, the average variability index, and the variability index of the management system for the previous time period, which is used for overall management of ride-hailing services. The expression is as follows: ； In the formula, is a management assignment, is a change degree index of the overall management of the system on the online car in the previous time period, is a change degree index average value, is a change degree index standard deviation, is a change degree threshold value, is a standard deviation threshold value; Management assignment of partition The larger, the more need to manage the partition first, get all the management assignment of partition Then, all the partition according to the management assignment Sort from large to small, generate management partition list; The calculation logic of the order change coefficient is as follows: the period when the number of user orders minus the number of ride-hailing vehicles exceeds the difference threshold is used as the period for order oversupply warning. During the monitoring period, the longer the period for order oversupply warning is, the more likely it is to lead to user churn. The period during which the number of ride-hailing vehicles minus the number of user orders exceeds a threshold is used as the period for order overrun warning. Within the monitoring period, the longer the period for order overrun warning is, the more likely it is to lead to the loss of ride-hailing drivers. The order surplus early warning period and the order overage early warning period are integrated to obtain an order change coefficient, and the calculation expression is: , wherein, is a real-time change amount of the partition order, is an order overage early warning period, is an order surplus early warning period.

2. The ride-hailing management method based on order analysis according to claim 1, characterized in that: The management system periodically obtains order data and various data related to ride-hailing operations from the ride-hailing management platform. The order data includes the order acceptance change coefficient, and the various data include the driver account cancellation rate, the duration deviation index, and the order completion rate.

3. The ride-hailing management method based on order analysis according to claim 1, characterized in that: Generating a variability index for each partition includes the following steps: The management system will input the order change coefficient, machine account cancellation rate, duration deviation index and order completion rate obtained periodically into the correlation model; The correlation model, after comprehensively analyzing the order change coefficient, machine account cancellation rate, time deviation index, and order completion rate, outputs a change index for each partition.

4. The ride-hailing management method based on order analysis according to claim 3, characterized in that: Determining whether a partition exhibits abnormal changes by comparing the change index with the change threshold includes the following steps: If the degree of change index is less than or equal to the degree of change threshold, it is determined that there is no abnormal change in the partition. If the change index is greater than the change threshold, it is determined that there is an abnormal change in the partition, and the management system sends a warning signal to the ride-hailing management platform.

5. The ride-hailing management method based on order analysis according to claim 1, characterized in that: The logic for obtaining the average value and standard deviation of the variability index is as follows: The time interval at which the management system periodically retrieves order data and various data related to ride-hailing operations from the ride-hailing management platform is marked as Td; The time interval for the management system to periodically conduct overall management of ride-hailing vehicles is marked as Tq; Then the number of historical change indexes obtained for the partition is n = Tq / Td-1: The expressions for calculating the mean and standard deviation of the variability index are as follows: ； In the formula, , This indicates the number of historical change indices for the partition. It is a positive integer. This represents the i-th degree of change index in the set of degree of change indices. This represents the average value of the degree of change index. It represents the standard deviation of the variability index.

6. The ride-hailing management method based on order analysis according to claim 5, characterized in that: After updating and sorting all partitions according to their degree of change from largest to smallest, generating an updated partition list includes the following steps: The obtained partition information is sorted using a sorting algorithm. Merge sort is used to sort the partition information from largest to smallest according to the degree of change. The sorted partition information is organized into an updated partition list, which contains information for each partition, including partition name, change index, and partition number. The updated list of partitions can be stored in a database or displayed directly.

7. A ride-hailing management system based on order analysis, used to implement the management method according to any one of claims 1-6, characterized in that: It includes a partition identification module, an initial sorting module, a data acquisition module, a data analysis module, an update sorting module, a judgment module, and a management sorting module. Zone Identification Module: Obtain the number of zones in the current city through the city management platform, and generate a unique identifier for each zone. Each identifier includes the zone name, zone economic status, and zone number. Initial sorting module: Performs an initial sorting of each partition based on economic conditions, generating an initial partition list; Data acquisition module: Periodically retrieves order data and various data related to ride-hailing operations from the ride-hailing management platform; Data Analysis Module: After analyzing order data and multiple data items in each partition using a correlation model, it generates a change index for each partition. Update sorting module: After updating and sorting all partitions according to the degree of change index from large to small, an updated partition list is generated; Judgment Module: The system determines whether there are any abnormal changes in a partition by comparing the change index with the change threshold. When an abnormal change is found in a partition, the system generates management suggestions for that partition and sends the updated sorting list and management suggestion information to the ride-hailing management platform. Management and sorting module: When performing overall management of ride-hailing services periodically, the module obtains the historical change index of each partition, analyzes the historical change index to generate a management value for each partition, sorts each partition from largest to smallest according to the management value, generates a management partition list, generates management strategies based on the management partition list, and sends them to the ride-hailing management platform.

Images