Designated driver user side order multi-dimensional state tracking and abnormity early warning system

Abstract

The invention relates to the technical field of computer software, in particular to a designated driver client order multi-dimensional state tracking and abnormity early warning system, which comprises a state acquisition module used for acquiring multi-dimensional state data of a designated driver order in real time; the multi-dimensional order tracking module is used for carrying out integration processing on the multi-dimensional state data, constructing an order full life cycle tracking model based on a space-time fusion algorithm, and generating a visual order state tracking view; the abnormity early warning module is used for identifying an order abnormity event, quantifying an abnormity level and generating abnormity early warning information; the order linkage processing module is used for linking a designated driving platform background and generating an optimal order adjustment strategy and an emergency disposal scheme by using an improved multi-target particle swarm optimization algorithm; and the user interaction module is used for pushing an order state tracking view, abnormal early warning information and a disposal scheme to the designated driver client, and providing a user feedback interaction channel. Therefore, the problems of state information splitting, abnormality discovery lagging and the like in the prior art are solved.

Description

Technical Field

[0001] This invention relates to the field of computer software technology, specifically to a multi-dimensional status tracking and anomaly warning system for ride-hailing user orders. Background Technology

[0002] Traditional ride-hailing order status management only provides a few simple status nodes such as "order accepted," "departed," "in service," and "completed." Users cannot grasp key information such as driver location dynamics, route deviations, and fare calculation progress in real time and in detail, making it difficult to form a clear understanding of the overall order progress. Current related technological developments mostly focus on optimizing single functions, such as improving map positioning accuracy and speeding up message push, but lack a comprehensive solution that can integrate multi-dimensional information such as location, route, fare, and driver status to achieve intelligent analysis and proactive early warning. As a result, users are always in a "passive information receiving" state during the service process and cannot detect potential problems in advance.

[0003] Existing ride-hailing apps with basic order status tracking functions have significant technical shortcomings: First, status information is fragmented; key data such as driver location, planned route, and real-time fare are presented independently, requiring users to switch between different interfaces to view them, making it difficult to quickly obtain a complete picture of the order. Second, anomaly detection is delayed; the system cannot identify anomalies such as driver lateness, route deviation, or fare exceeding the estimate in real time, relying on users' subjective observation to discover problems. Furthermore, users can only submit complaints through the "Complaints and Feedback" page after the service is completed, lacking an in-process intervention mechanism, leading to the continued escalation of the impact of anomalies. Third, users face difficulties in providing evidence; when appealing, they can only provide subjective information such as text descriptions, lacking objective evidence such as automatically recorded trajectory data and fare details. This not only reduces customer service efficiency but also easily leads to disputes between users, the platform, and drivers due to insufficient evidence. Fourth, the interaction channels are limited; the feedback entry point is buried deep within multiple levels of pages such as "My - Settings - Help and Feedback," making the process complex and making it difficult for users to quickly initiate feedback when an anomaly occurs, further exacerbating the poor service experience. Summary of the Invention

[0004] This application provides a multi-dimensional status tracking and anomaly warning system for ride-hailing user orders to solve the problems of fragmented status information and delayed anomaly detection in the prior art.

[0005] The first aspect of this application provides a multi-dimensional status tracking and anomaly warning system for chauffeur service user orders, comprising: a status acquisition module, a multi-dimensional order tracking module, an anomaly warning module, an order linkage processing module, and a user interaction module; wherein, the status acquisition module is used to collect multi-dimensional status data of chauffeur service orders in real time; the multi-dimensional order tracking module is used to integrate and process the multi-dimensional status data, construct an order lifecycle tracking model based on a spatiotemporal fusion algorithm, and generate a visualized order status tracking view; the anomaly warning module is used to monitor changes in order status in real time, identify abnormal events in orders and quantify the anomaly level by combining preset multi-dimensional anomaly judgment thresholds, and generate anomaly warning information; the order linkage processing module is used to receive the warning information, link with the chauffeur service platform backend, and use an improved multi-objective particle swarm optimization algorithm to generate an optimal order adjustment strategy and emergency response plan; the user interaction module is used to push the order status tracking view, anomaly warning information, and response plan to the chauffeur service user terminal, and provide a user feedback and interaction channel.

[0006] Preferably, the status acquisition module includes an order basic information acquisition unit, a driver status acquisition unit, a trip environment acquisition unit, and a vehicle status acquisition unit. The order basic information acquisition unit is used to collect static data such as order number, departure point, destination, and appointment time, and to update the current order progress in real time. The driver status acquisition unit is used to simultaneously collect the driver's real-time location, driving speed, and the order acceptance status, service status, and real-time communication status fed back by the driver's terminal. The trip environment acquisition unit is used to collect information such as congestion level, weather conditions, road construction information, and traffic control information for the trip route.

[0007] Preferably, the multi-dimensional order tracking module includes a data integration unit, a spatiotemporal fusion tracking model construction unit, and a visualization view generation unit. The data integration unit performs deduplication, noise reduction, and format standardization on multi-dimensional status data to generate a unified dataset and generates an immutable snapshot of the entire order lifecycle based on timestamps. The spatiotemporal fusion tracking model construction unit processes spatiotemporal data using a spatiotemporal graph convolutional network with an improved attention mechanism to construct an order lifecycle tracking model, mapping the order status in real time from creation, order acceptance, pick-up, trip, to completion. The visualization view generation unit generates a visualization interface that includes a real-time trajectory map, a status timeline, key node countdowns, a multi-dimensional data dashboard, and a data snapshot backtracking entry, supporting user zooming and backtracking.

[0008] Preferably, the anomaly warning module includes an anomaly rule storage unit, a real-time status monitoring unit, an anomaly identification unit, a level quantification unit, and an early warning generation unit; wherein, the anomaly rule storage unit is used to pre-store multi-dimensional anomaly judgment thresholds for time delays, location deviations, driver anomalies, and resource shortages; the real-time status monitoring unit is used to extract dynamic status data related to anomalies from the order lifecycle tracking model every second, compare it with the multi-dimensional anomaly judgment thresholds, and capture anomaly symptoms; the anomaly identification unit, based on the anomaly symptoms, analyzes the cause of the anomaly through a feature matching algorithm to identify the specific anomaly type; the level quantification unit is used to classify anomalies into three levels: minor, moderate, and severe, and generate anomaly warning information containing the cause of the anomaly, the anomaly type, the scope of impact, and risk warnings.

[0009] Preferably, the order linkage processing module includes an order anomaly handling strategy generation unit and a platform linkage unit. The order anomaly handling strategy generation unit receives the warning information, extracts the anomaly type, level, and impact range, and, based on an improved multi-objective particle swarm optimization algorithm, generates order adjustment strategies for minor and moderate warnings with the objective functions of "minimizing delay time, maximizing user satisfaction, and minimizing scheduling costs." For severe warnings, it simultaneously generates order adjustment strategies and emergency response plans. The platform linkage unit connects with the ride-hailing platform backend via an API interface to synchronize order adjustment strategies, emergency response plans, execution status, and associated order data snapshots.

[0010] Preferably, the user interaction module includes a view push unit, an early warning and solution push unit, and a feedback interaction unit. The view push unit displays the order status tracking view in real-time on the homepage and order details page of the chauffeur service user app, and supports quick viewing via a floating window. The early warning and solution push unit uses a multi-channel push mode, including app pop-ups, SMS, and voice prompts, prioritizing the push of high-level early warning information and corresponding handling solutions. The feedback interaction unit provides interactive functions such as one-click confirmation of handling solutions, refusal of adjustments, consultation with customer service, and customized feedback. User feedback information is synchronized to the platform backend in real-time, supporting dynamic strategy optimization.

[0011] The second aspect of this application provides a method for a multi-dimensional status tracking and anomaly warning system for ride-hailing user-end orders, comprising: acquiring multi-dimensional status data of ride-hailing orders; integrating and processing the multi-dimensional status data of the ride-hailing orders, generating an immutable data snapshot of the entire lifecycle of the order based on a timestamp, and simultaneously constructing an order lifecycle tracking model using a spatiotemporal fusion algorithm to generate a visualized order status tracking view; extracting dynamic status data related to anomalies from the order lifecycle tracking model, comparing it with anomaly judgment thresholds, analyzing the causes of anomalies using a feature matching algorithm, identifying abnormal events in the order and quantifying the anomaly level, and generating anomaly warning information; based on the warning information, linking with the ride-hailing platform backend, using an improved multi-objective particle swarm optimization algorithm to generate an optimal order adjustment strategy and emergency response plan; displaying the visualized order status tracking view, anomaly warning information, order adjustment strategy, and emergency response plan on the user end, and providing a user feedback channel.

[0012] A third aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the program to implement a method for a multi-dimensional status tracking and anomaly warning system for chauffeur service user orders as described in the above embodiments.

[0013] The fourth aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which is executed by a processor to implement a method for a multi-dimensional status tracking and anomaly warning system for chauffeur service user orders as described in the above embodiments.

[0014] The fifth aspect of this application provides a computer program product, including a computer program or instructions, for implementing a method for a multi-dimensional status tracking and anomaly warning system for chauffeur service user orders as described in the above embodiments.

[0015] Therefore, this application has the following beneficial effects: This application embodiment comprehensively collects multi-dimensional data on basic order information, driver status, and trip environment through a status acquisition module. This breaks through the limitations of traditional single-dimensional data collection, providing complete and accurate basic data support for order tracking and anomaly identification, avoiding tracking deviations and missed warnings due to missing information. The multi-dimensional order tracking module performs deduplication, noise reduction, and standardization on the data. It constructs a full lifecycle tracking model based on a spatiotemporal graph convolutional network with an improved attention mechanism, generating a visual view containing real-time trajectories and timelines. This allows users to intuitively grasp the status of each stage of the order and supports snapshot backtracking, solving the problem of traditional tracking lacking spatiotemporal correlation and visualization. The anomaly warning module monitors anomalies in real time by preset multi-dimensional anomaly thresholds. It captures abnormal signs, analyzes the causes using feature matching algorithms, quantifies three levels of anomaly, and accurately identifies and warns of abnormal events in advance, avoiding risks caused by delayed anomaly detection. The order linkage processing module works with the platform backend, using an improved multi-objective particle swarm optimization algorithm to generate optimal adjustment strategies and contingency plans, balancing delay time, user satisfaction, and scheduling costs, and improving the efficiency and rationality of anomaly handling. The user interaction module provides diverse feedback channels by pushing views, warnings, and solutions through multiple channels, ensuring that users can obtain information and participate in interactions in a timely manner. Synchronous feedback data supports dynamic strategy optimization, significantly improving the intelligence level of ride-hailing order management, anomaly response speed, and user experience, while reducing order risks and scheduling costs. Thus, it solves the problems of fragmented state information and delayed anomaly detection in existing technologies.

[0016] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0017] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a schematic diagram of the structure of a multi-dimensional status tracking and anomaly warning system for chauffeur service user orders provided according to an embodiment of this application; Figure 2 This is a schematic diagram of a status acquisition module provided according to an embodiment of this application; Figure 3 This is a schematic diagram of a multi-dimensional order tracking module provided according to an embodiment of this application; Figure 4 This is a schematic diagram of an anomaly warning module provided according to an embodiment of this application; Figure 5 This is a schematic diagram of an order linkage processing module provided according to an embodiment of this application; Figure 6This is a schematic diagram of a user interaction module provided according to an embodiment of this application; Figure 7 A flowchart of a method for a multi-dimensional status tracking and anomaly warning system for chauffeur service user orders provided in an embodiment of this application; Figure 8 This is a schematic diagram of a method for multi-dimensional status tracking and anomaly warning system for chauffeur service user orders according to an embodiment of this application; Figure 9 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of this application. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0019] The following description, with reference to the accompanying drawings, illustrates a multi-dimensional status tracking and anomaly warning system for ride-hailing user-side orders according to an embodiment of this application. Addressing the issue of delayed anomaly detection mentioned in the background section, this application provides a multi-dimensional status tracking and anomaly warning system for ride-hailing user-side orders. In this system, a status acquisition module comprehensively collects multi-dimensional data on basic order information, driver status, and travel environment, breaking the limitations of traditional single-dimensional data collection. This provides complete and accurate basic data support for order tracking and anomaly identification, avoiding tracking deviations and missed warnings due to missing information. The multi-dimensional order tracking module performs deduplication, noise reduction, and standardization on the data. It constructs a full lifecycle tracking model based on a spatiotemporal graph convolutional network with an improved attention mechanism, generating a visual view containing real-time trajectories, timelines, etc., allowing users to intuitively grasp the status of each stage of the order. It supports snapshot backtracking, solving the problem of traditional tracking lacking spatiotemporal correlation and visualization. The system addresses several issues. The anomaly warning module, by pre-setting multi-dimensional anomaly thresholds, monitors and captures anomaly signs in real time. It uses feature matching algorithms to analyze causes and quantify three levels of anomaly severity, enabling accurate identification and early warning of abnormal events, thus mitigating risks caused by delayed anomaly detection. The order linkage processing module, in conjunction with the platform backend, utilizes an improved multi-objective particle swarm optimization algorithm to generate optimal adjustment strategies and contingency plans, balancing delay time, user satisfaction, and scheduling costs, thereby improving the efficiency and rationality of anomaly handling. The user interaction module provides diverse feedback channels by pushing views, warnings, and solutions through multiple channels, ensuring users receive information promptly and participate in interactions. Synchronous feedback data supports dynamic strategy optimization, significantly improving the intelligence level of ride-hailing order management, anomaly response speed, and user experience, while reducing order risks and scheduling costs. This solves the problems of fragmented state information and delayed anomaly detection in existing technologies.

[0020] Figure 1 This is a schematic diagram of the structure of a multi-dimensional status tracking and anomaly warning system for chauffeur service user orders provided in an embodiment of this application.

[0021] This application provides a multi-dimensional status tracking and anomaly warning system for chauffeur service user orders. The system 10 includes: a status acquisition module 100, a multi-dimensional order tracking module 200, an anomaly warning module 300, an order linkage processing module 400, and a user interaction module 500. The system comprises the following modules: Status Acquisition Module 100, which collects multi-dimensional status data of chauffeur orders in real time; Multi-dimensional Order Tracking Module 200, which integrates and processes the multi-dimensional status data, constructs a full lifecycle tracking model for orders based on a spatiotemporal fusion algorithm, and generates a visual order status tracking view; Anomaly Warning Module 300, which monitors changes in order status in real time, identifies abnormal events and quantifies the anomaly level by combining preset multi-dimensional anomaly judgment thresholds, and generates anomaly warning information; Order Linkage Processing Module 400, which receives warning information, links with the chauffeur platform backend, and uses an improved multi-objective particle swarm optimization algorithm to generate optimal order adjustment strategies and emergency response plans; and User Interaction Module 500, which pushes order status tracking views, anomaly warning information, and response plans to chauffeur users, and provides a user feedback and interaction channel.

[0022] It is understood that in this embodiment, the status acquisition module comprehensively collects multi-dimensional data on basic order information, driver status, and trip environment, breaking the limitations of traditional single-dimensional data collection. This provides complete and accurate basic data support for order tracking and anomaly identification, avoiding tracking deviations and missed warnings due to missing information. The multi-dimensional order tracking module performs deduplication, noise reduction, and standardization on the data. It builds a full lifecycle tracking model based on a spatiotemporal graph convolutional network with an improved attention mechanism, generating a visual view containing real-time trajectory, timeline, etc., allowing users to intuitively grasp the status of each stage of the order. It supports snapshot backtracking, solving the problem of traditional tracking lacking spatiotemporal correlation and visualization. The anomaly warning module presets multi-dimensional anomaly thresholds. Real-time monitoring and capture of abnormal signs, using feature matching algorithms to analyze causes and quantify three levels of anomaly, enables accurate identification and early warning of abnormal events, mitigating risks caused by delayed anomaly detection. The order linkage processing module, in conjunction with the platform backend, utilizes an improved multi-objective particle swarm optimization algorithm to generate optimal adjustment strategies and contingency plans, balancing delay time, user satisfaction, and scheduling costs, thereby improving the efficiency and rationality of anomaly handling. The user interaction module provides diverse feedback channels by pushing views, warnings, and solutions through multiple channels, ensuring users receive information and participate in interactions in a timely manner. Synchronous feedback data supports dynamic strategy optimization, significantly improving the intelligence level of ride-hailing order management, anomaly response speed, and user experience, while reducing order risks and scheduling costs. This solves the problems of fragmented state information and delayed anomaly detection in existing technologies.

[0023] In this embodiment of the application, the status acquisition module 100 includes: as follows Figure 2 As shown, there are three units: the order basic information collection unit, the driver status collection unit, and the trip environment collection unit.

[0024] The order basic information collection unit is used to collect static data such as order number, departure point, destination, and appointment time, and update the current order progress in real time; the driver status collection unit is used to collect the driver's real-time location, driving speed, and the order acceptance status, service status, and real-time communication status fed back by the driver's terminal; the trip environment collection unit is used to collect the congestion level, weather conditions, road construction information, and traffic control information of the trip route.

[0025] It is understood that the order basic information collection unit in this application embodiment establishes a unique identifier and basic information file for each chauffeur order by collecting static data such as order number and departure location, and updates the order progress in real time, so that users and the platform can simultaneously grasp the stage of the order (such as order acceptance, pick-up, trip, etc.), avoiding misunderstandings caused by information asynchrony; the chauffeur status collection unit simultaneously captures the real-time location, driving speed, and order acceptance, service, and communication status of the chauffeur, which can not only accurately locate the chauffeur's dynamics to estimate the pick-up time, but also determine whether there are communication obstacles through the communication status, and avoid service connection problems in advance; the trip environment collection unit obtains congestion level, weather, construction and control information, which can predict the risk of trip delays and provide environmental dimension basis for subsequent abnormal warnings; the collaboration of multiple units realizes full-chain data coverage of the order, providing comprehensive and accurate basic data support for subsequent multi-dimensional tracking and abnormal identification.

[0026] For example, after a business dinner in a commercial area, Ms. Wang booked a designated driver service through a designated driver app from the underground parking garage of "Xingguang Tiandi Shopping Mall" to "Lvcheng Jiayuan Community". The moment the order was successfully placed, the basic order information collection unit immediately collected static data such as the order number "DJ202406152110", the precise location of the departure point (latitude and longitude coordinates), the detailed address of the destination, and the appointment time "21:10". Subsequently, as the service progressed, the order progress was updated in real time as "Pending order → Driver accepted order → Driver on the way to pick up → Arrived at pick-up point → In progress → Approaching destination", allowing Ms. Wang to clearly see each key milestone on the app. After driver Mr. Zhao accepted the order, the driver status collection unit simultaneously captured his real-time location (0.8 km from the pick-up point) and driving speed (28 km / h), and his handheld terminal provided feedback on the order status as "Confirmed", the service status as "Pick-up in progress", and the real-time communication status as "Online and available for calls". Ms. Wang could intuitively track Mr. Zhao's movement trajectory through the app's built-in map, estimating that he would arrive in 5 minutes. At this time, the environmental monitoring unit detected that Jiefang Avenue was experiencing moderate congestion due to the end of the evening rush hour, the current weather was "sunny turning to light breeze," there was nighttime road maintenance work on Yuanding Road (occupying one lane), and there was no traffic control on the Ring Road. This information was synchronized to the platform in real time, providing a reference for the estimated travel time. Throughout the journey, each monitoring unit continuously updated the data. Ms. Wang could check the trip progress, the driver's status, and the vehicle's real-time condition through the app. The platform dynamically adjusted the estimated delivery time based on this data, ensuring the transparency and safety of the chauffeur service.

[0027] In this embodiment of the application, the multi-dimensional order tracking module 200 includes: Figure 3 As shown, there are three units: data integration unit, spatiotemporal fusion tracking model construction unit, and visualization view generation unit.

[0028] The data integration unit is used to deduplicate, reduce noise, and standardize the format of multi-dimensional status data to generate a unified dataset and generate an immutable data snapshot of the entire order lifecycle based on timestamps. The spatiotemporal fusion tracking model construction unit is used to process spatiotemporal data based on a spatiotemporal graph convolutional network with an improved attention mechanism to build an order lifecycle tracking model, which maps the status of each stage of the order from creation, order acceptance, pick-up, trip to completion in real time. The visualization view generation unit is used to generate a visualization interface that includes a real-time trajectory map, status timeline, key node countdown, multi-dimensional data dashboard, and data snapshot backtracking entry, supporting users to zoom and backtrack.

[0029] It is understood that the data integration unit in this application performs deduplication, noise reduction, and format standardization on multi-dimensional state data to generate a unified dataset and create an immutable data snapshot, ensuring the authenticity and integrity of order data and providing a reliable basis for subsequent traceability. The spatiotemporal fusion tracking model construction unit relies on a spatiotemporal graph convolutional network with an improved attention mechanism to accurately process spatiotemporal data. The constructed full lifecycle tracking model can map the status of each stage of the order in real time and clearly present the dynamic evolution trajectory of the order. The visualization view generation unit generates real-time trajectory maps, status timelines, and other visualization interfaces, which support zooming and backtracking. This allows users to intuitively grasp the core information of the order and facilitates the platform to quickly verify order details, effectively solving the problems of fragmented traditional order tracking data and low visualization, and improving the transparency and efficiency of order management.

[0030] It should be noted that the spatiotemporal fusion algorithm refers to an algorithm that integrates data from both temporal and spatial dimensions (such as order progress and time points) to build a model that enables dynamic tracking and visualization of the entire order lifecycle. The formula is:

[0031]

[0032]

[0033]

[0034]

[0035] in, For time The node feature matrix at time step; This is the normalized graph adjacency matrix; For trainable convolutional kernel weights; For bias terms; For activation functions; This is the output feature matrix after spatial convolution; This is a one-dimensional convolution operation; The size of the convolution kernel over time; For the past The input feature matrix at each time step; These are the weights of the temporal convolution kernel; This refers to the temporal convolution bias term; This is the output feature matrix after temporal convolution; For nodes For nodes Attention weights; For nodes eigenvectors; For nodes eigenvectors; A function for scoring the correlation of node features; For nodes All related node indexes; Attention-weighted nodes Feature vector; For nodes For nodes Attention weights; for Output results of the real-time order lifecycle tracking model; It is a spatiotemporal graph convolutional network.

[0036] Formulas for generating a visual interface:

[0037]

[0038]

[0039]

[0040] in, The screen X coordinates of the target location on the map control; This refers to the width of the map control. The GPS longitude of the target location; The minimum longitude of the currently visible area on the map; This represents the maximum longitude of the currently visible area on the map. The screen Y-coordinate of the target location on the map control; This refers to the height of the map control. The GPS latitude of the target location; This represents the maximum latitude of the currently visible map area. The minimum latitude of the currently visible map area; The remaining time for the critical nodes; For the expected timestamps of key nodes; This refers to the rotation angle of the instrument panel pointer; This is the minimum rotation angle of the instrument panel pointer; This represents the maximum rotation angle of the instrument panel pointer. This is the current monitoring value; The minimum range for monitoring values; The maximum range for monitoring values.

[0041] For example, after Mr. Li booked a chauffeur service from his office building to his residential area, the data integration unit immediately began its work: first, it deduplicated (removing three duplicate location data entries uploaded by the driver's terminal) and reduced noise (filtering out two abnormal fluctuations in tire pressure sensor values ​​caused by road bumps) from multiple sources of data, including basic order information, driver location, trip environment, and vehicle status; and standardized data such as latitude and longitude and fuel consumption units in different formats to generate a unified dataset. Simultaneously, it generated immutable data snapshots based on timestamps every 10 seconds, showing the sequence "order creation → order acceptance → pick-up → trip in progress." The spatiotemporal fusion tracking model construction unit, relying on a spatiotemporal graph convolutional network with an improved attention mechanism, processed this spatiotemporal data, constructing a full lifecycle tracking model that maps the order status in real time and accurately correlates the driver's driving trajectory with the trip progress. The visualization view generation unit generates a customized interface for Mr. Li's APP: the real-time trajectory map clearly displays the real-time location and driving route of the designated driver, the status timeline marks the completion time of each node, the countdown prompt for key nodes shows "expected arrival at the destination in 15 minutes", the multi-dimensional data dashboard synchronously displays the vehicle's tire pressure, fuel consumption and road congestion level, and there is also a data snapshot retrospective entry. Mr. Li can zoom in and out of the map to view details, and can also retrospectively review the order data at any point in time, so as to intuitively grasp the service dynamics throughout the process. In this embodiment of the application, the anomaly warning module 300 includes: Figure 4 As shown, the system includes an anomaly rule storage unit, a real-time status monitoring unit, an anomaly identification unit, a level quantification unit, and an early warning generation unit.

[0042] The system includes: an anomaly rule storage unit for pre-storing multi-dimensional anomaly judgment thresholds for time delays, location deviations, driver abnormalities, and resource shortages; a real-time status monitoring unit for extracting dynamic status data related to anomalies from the order lifecycle tracking model every second, comparing it with the multi-dimensional anomaly judgment thresholds, and capturing anomaly signs; an anomaly identification unit for analyzing the causes of anomalies and identifying specific anomaly types based on anomaly signs and through feature matching algorithms; and a level quantification unit for classifying anomalies into three levels: minor, moderate, and severe, and generating anomaly warning information containing the cause of the anomaly, the type of anomaly, the scope of impact, and risk warnings.

[0043] It is understood that the embodiments of this application pre-store multi-dimensional anomaly judgment thresholds through the anomaly rule storage unit, providing clear standards for anomaly identification and avoiding subjective judgment bias; the real-time status monitoring unit extracts dynamic status data every second and compares it with the thresholds, which can quickly capture abnormal signs such as time delays and location deviations, and realize early perception of anomalies; the anomaly identification unit analyzes the cause of anomalies and identifies specific types through feature matching algorithms, providing direction for precise handling; the level quantification unit divides anomalies into three levels and generates early warning information containing key information, which not only allows users and the platform to clearly know the severity of the anomaly, but also allows high-risk events to be handled first according to the level, avoiding order disputes or safety hazards caused by untimely handling of anomalies, and improving the reliability of the chauffeur service.

[0044] For example, Ms. Chen booked a chauffeur service from the high-speed rail station to a villa area. The anomaly rule storage unit pre-stored multi-dimensional anomaly judgment thresholds such as "route deviation exceeding 500 meters" and "pickup delay exceeding 15 minutes." During the trip, the real-time status monitoring unit extracted dynamic order data every second and found that the chauffeur's driving trajectory deviated from the planned route by 600 meters, and the scenic avenue along the route experienced a sudden traffic accident, causing congestion to rise to "severe," triggering anomaly signs. The anomaly identification unit analyzed the data using a feature matching algorithm and determined the anomaly type to be "route deviation + trip delay," caused by a sudden traffic event that forced the chauffeur to take a temporary detour. The level quantification unit classified it as a "moderate anomaly," and the warning generation unit immediately generated a warning message containing "the cause of the anomaly, the scope of impact, and the estimated delay of 8 minutes," which was pushed to the platform and Ms. Chen's device in real time, proactively avoiding user dissatisfaction caused by lack of awareness.

[0045] In this embodiment of the application, the order linkage processing module 400 includes: Figure 5 As shown, there are two units: the order exception handling strategy generation unit and the platform linkage unit.

[0046] The order anomaly handling strategy generation unit receives early warning information, extracts the anomaly type, level, and scope of impact, and generates order adjustment strategies for minor and moderate early warnings based on an improved multi-objective particle swarm optimization algorithm with the objective functions of "minimizing delay time, maximizing user satisfaction, and minimizing scheduling costs." For severe early warnings, it simultaneously generates order adjustment strategies and emergency response plans. The platform linkage unit connects with the chauffeur platform backend through an API interface to synchronize order adjustment strategies, emergency response plans, execution status, and associated order data snapshots.

[0047] It is understood that the order exception handling strategy generation unit in this application extracts key exception information and, based on an improved multi-objective particle swarm optimization algorithm, generates targeted adjustment strategies and emergency plans for different levels of exceptions with the goals of "minimizing delays, maximizing satisfaction, and minimizing costs," ensuring that the handling plans are scientific and reasonable. The platform linkage unit links with the backend through API interfaces to synchronize strategies, plans, and data snapshots, enabling rapid implementation of order adjustments and emergency handling, avoiding delays caused by information fragmentation. This module can respond quickly to exceptions without manual intervention, which not only improves the efficiency of exception handling but also reduces scheduling costs and user dissatisfaction rates through optimized strategies, ensuring the continuity and stability of the chauffeur service.

[0048] It should be noted that the improved multi-objective particle swarm optimization algorithm refers to an upgraded version of the particle swarm optimization algorithm, with the core objectives of "minimizing delay time, maximizing user satisfaction, and minimizing scheduling costs." This algorithm is used to quickly solve for optimal order adjustment strategies and emergency response plans for different levels of anomalies in ride-hailing orders. The formula is:

[0049]

[0050]

[0051] in, This is the scheduling decision vector; For order delay duration; For user satisfaction; (For the reverse target of satisfaction) For scheduling costs; Optimize vectors for multiple objectives; For the first The generation A vector of particle velocities; Inertial weight; For the first The generation The velocity of each particle; The learning factor is set to 2 to guide particles to learn towards their optimal learning path. The learning factor is set to 2 to guide particles toward the optimal learning path of the population. The first random number in the interval [0,1]; The second random number in the interval [0,1]; For the first The optimal position of each individual particle; For the first The generation The position of each particle; This represents the globally optimal position for the particle swarm. For the first The generation The position of each particle.

[0052] For example, on Saturday evening, Ms. Zhao booked a designated driver to take her from a restaurant in the south of the city to her residential area in the north. After accepting the order, driver Mr. Sun suddenly experienced acute stomach pain and was unable to continue the service while driving halfway. He immediately reported a "service interruption" anomaly through his terminal. The order linkage processing module responded immediately: the order anomaly handling strategy generation unit extracted the key information of "serious anomaly, service interruption" and, based on an improved multi-objective particle swarm optimization algorithm, generated the optimal strategy with the goal of "fastest service connection and minimal user waiting time"—instructing Mr. Sun to park the user's vehicle in a safe parking lot nearby, and simultaneously dispatching the nearest (1.8 km) standby driver, Mr. Liu, to take over the order, with an estimated arrival time of 12 minutes; the platform linkage unit deeply integrated with the designated driver platform backend through the API interface, synchronously creating the relay order, pushing Mr. Liu's real-time location and contact information to Ms. Zhao and Mr. Sun's terminals, and the backend automatically updated the order status to "anomaly handling → relay service in progress," and retained a snapshot of the current order data. After Mr. Liu arrived on time, the trip was seamlessly connected. In the end, Ms. Zhao only had to wait 10 minutes longer before arriving home safely. No additional action was required from the user, ensuring service continuity and user experience.

[0053] In this embodiment of the application, the user interaction module 500 includes: Figure 6 As shown, there are a view push unit, an early warning and solution push unit, and a feedback interaction unit.

[0054] The view push unit displays the order status tracking view in real time on the homepage and order details page of the chauffeur user's APP, and supports quick viewing via a floating window; the warning and solution push unit uses a multi-channel push mode, including APP pop-ups, SMS, and voice prompts, to prioritize the push of high-level warning information and corresponding handling solutions; the feedback interaction unit provides interactive functions such as one-click confirmation of handling solutions, refusal of adjustments, consultation with customer service, and customized feedback. User feedback information is synchronized to the platform backend in real time to support dynamic strategy optimization.

[0055] It is understood that the view push unit in this application embodiment will display the tracking view in real time on the APP homepage and details page and support floating window viewing, so that users can quickly obtain the order status at any time without cumbersome operations; the warning and solution push unit adopts a multi-channel push mode, prioritizing the push of high-level warnings and handling solutions, ensuring that users know about abnormal situations and solutions as soon as possible; the feedback interaction unit provides diversified interactive functions, and user feedback information is synchronized to the backend in real time, which not only meets the user's needs for independent selection and consultation, but also provides real user data support for the platform to dynamically optimize order strategies, effectively improving user participation and service experience, and reducing complaints caused by information asymmetry or lack of feedback channels.

[0056] For example, after Mr. Li booked a designated driver, the user interaction module played a crucial role throughout the process: the view push unit displayed the order status tracking view in real time on the order details page of the APP, and the floating window allowed for quick viewing of the designated driver's real-time trajectory and vehicle status; when a minor congestion warning occurred during the trip, the warning and solution push unit pushed the information "Road congestion, estimated delay of 5 minutes, route optimized" through both APP pop-ups and SMS, prioritizing highly relevant content; if Mr. Li had questions about the new route, he could communicate in real time through the "Consult Customer Service" function of the feedback interaction unit, and customer service would simultaneously retrieve the order data snapshot to answer the questions. After the communication, Mr. Li clicked "Confirm Adjustment," and the feedback information was synchronized to the platform backend in real time, providing user behavior data support for subsequent optimization of the route planning algorithm. The entire process was convenient and the information was transparent.

[0057] This application proposes a multi-dimensional status tracking and anomaly warning system for ride-hailing user orders. Through a status acquisition module, it comprehensively collects multi-dimensional data on basic order information, driver status, and travel environment, breaking the limitations of traditional single-dimensional data collection. This provides complete and accurate basic data support for order tracking and anomaly identification, avoiding tracking deviations and missed warnings due to missing information. The multi-dimensional order tracking module performs deduplication, noise reduction, and standardization on the data. It constructs a full lifecycle tracking model based on a spatiotemporal graph convolutional network with an improved attention mechanism, generating a visual view containing real-time trajectories and timelines. This allows users to intuitively grasp the status of each stage of the order and supports snapshot backtracking, solving the problem of traditional tracking lacking spatiotemporal correlation and visualization. The anomaly warning module... By pre-setting multi-dimensional anomaly thresholds, the system monitors and captures anomaly signs in real time. It utilizes feature matching algorithms to analyze causes and quantify three levels of anomaly severity, enabling accurate identification and early warning of abnormal events, thus mitigating risks caused by delayed anomaly detection. The order linkage processing module, in conjunction with the platform backend, uses an improved multi-objective particle swarm optimization algorithm to generate optimal adjustment strategies and contingency plans, balancing delay time, user satisfaction, and scheduling costs to improve the efficiency and rationality of anomaly handling. The user interaction module provides diverse feedback channels by pushing views, warnings, and solutions through multiple channels, ensuring users receive information and participate in interactions promptly. Synchronous feedback data supports dynamic strategy optimization, significantly improving the intelligence level of ride-hailing order management, anomaly response speed, and user experience, while reducing order risks and scheduling costs. This solves the problems of fragmented state information and delayed anomaly detection in existing technologies.

[0058] The following will illustrate a multi-dimensional status tracking and anomaly warning system for ride-hailing user orders through a specific embodiment, including: At 9 p.m. on Friday, outside the parking lot of the "Yuexuan Pavilion" restaurant in the city center, Mr. Wang had just finished a business engagement. He took out his phone, opened the designated driver app, entered the restaurant parking lot as the departure point and the "Yunxi Garden" residential area in the suburbs as the destination, and submitted the order after confirming that the vehicle was a black sedan. The moment an order is generated, the system's status acquisition module begins operating: the order basic information acquisition unit records the order number DJ2024051709001, the appointment time as immediate, and the estimated trip distance as 32 kilometers, while simultaneously updating the order progress to "pending order" in real time; the driver status acquisition unit filters nearby available drivers in the background and quickly matches Mr. Li, a driver located 3.1 kilometers from the restaurant in the parking lot of the adjacent shopping mall, and simultaneously collects his real-time location, the battery level of his electric vehicle (60% remaining), the "pending order" status feedback from the terminal, and the information that the communication signal is full; the trip environment acquisition unit captures static data such as that the evening rush hour has passed, the congestion level of the main roads around the restaurant is "moderate" (average speed 25km / h), the weather is cloudy and rainless, and there is no road construction or traffic control along the way. These multi-dimensional status data are transmitted in real time to the multi-dimensional order tracking module. The data integration unit first deduplicates the duplicate location information and filters out the occasional fluctuating signal interference data, generates a dataset in a unified format, and then generates the first order snapshot based on the timestamp. The spatiotemporal fusion tracking model construction unit combines Mr. Li's movement trajectory with the order route through a spatiotemporal graph convolutional network with an improved attention mechanism to build a full life cycle tracking model, mapping the complete stages of "order creation - order acceptance - pick-up - in progress - completion". The visualization view generation unit generates tracking views on Mr. Wang's APP homepage and order details page: Mr. Li's real-time location is marked with blue dots on the map, the planned route is marked with red routes, and key nodes such as "order acceptance (09:01) - expected pick-up (09:16) - expected arrival (09:48)" are clearly marked on the timeline. The dashboard also displays the congestion level and the driver's driving speed (currently 20km / h). Mr. Wang can click the "floating window" button next to the profile picture to view this information at any time.

[0059] After receiving the order notification on the terminal, Mr. Li immediately clicked "Confirm Order" and rode his electric bike towards the restaurant. The first five minutes went smoothly; he traveled 1.8 kilometers along the secondary road. Just as he turned onto the intersection leading to the restaurant, the electric bike suddenly lost power—the previously displayed 60% battery level was actually a false alarm caused by a delay in the terminal's sensors, rendering the vehicle completely unusable. Mr. Li immediately clicked "Report Anomaly" on the terminal, selected "Vehicle Malfunction," and attached photos of the scene. At this moment, the real-time status monitoring unit of the anomaly warning module was extracting data from the order tracking model every second. It detected that Mr. Li's location at the intersection had been stagnant for more than one minute, and his speed had plummeted from 20 km / h to 0. This perfectly matched the judgment threshold of "stagnant location for more than one minute during pick-up, speed suddenly drops" in the anomaly rule storage unit, and the anomaly was immediately detected. The anomaly identification unit, through a feature matching algorithm, combined Mr. Li's "fault report" information with location and speed data, quickly analyzing the cause of the anomaly as "fault of the designated driver's electric vehicle," ruling out the possibility of a traffic accident or deliberate delay. The level quantification unit, based on the situation that "the fault point is 2.3 kilometers away from the user, the re-dispatch of the designated driver is expected to cause a delay of 10-15 minutes, and there is no safety risk," set the anomaly level as "medium," and finally generated a warning message containing "Cause of anomaly: Fault of the designated driver Mr. Li's electric vehicle; Type of anomaly: Abnormal pick-up resources; Scope of impact: Delay in pick-up time; Risk warning: The designated driver needs to be replaced to reduce waiting time." Upon receiving the alert, the order linkage processing module immediately extracts key information about the anomaly and initiates an improved multi-objective particle swarm optimization algorithm. The objective functions are "minimizing delay time, maximizing user satisfaction, and minimizing scheduling cost." In the backend database, driver Zhang, who is 1.2 kilometers away from the restaurant, has just completed his previous order. His terminal shows that he is available and his electric vehicle has sufficient battery power. The estimated time to reach the restaurant is only 8 minutes, which is at least 5 minutes shorter than dispatching other drivers 3 kilometers away. The scheduling cost is also lower, which meets the optimal solution of the three objective functions. At the same time, an emergency plan is generated—if driver Zhang encounters obstacles on the way, backup driver Liu is on standby 2 kilometers away to ensure that there will be no further delays. The platform linkage unit synchronizes data with the chauffeur platform's backend via API, transmitting adjustment strategies and emergency plans to relevant terminals in real time. The user interaction module immediately pushes information to Mr. Wang via APP pop-up and SMS: "Hello! Your chauffeur, Mr. Li, encountered an electric vehicle malfunction during the pick-up. The system has replaced you with chauffeur, Mr. Zhang, who is expected to arrive at the restaurant parking lot in 8 minutes. If you have any questions, please contact customer service." Mr. Wang clicks the "Confirm Adjustment" button in the pop-up, and the feedback information is synchronized to the platform backend in real time. The system then sends a reminder to Mr. Li that "the order has been transferred, please pay attention to safety" and sends a dispatch notification containing Mr. Wang's vehicle information and location to Mr. Zhang.

[0060] Eight minutes later, Mr. Zhang arrived at the restaurant parking lot on time on his electric bike. He scanned the order QR code provided by Mr. Wang through the terminal to confirm the pick-up information. At this time, the status acquisition module updated Mr. Zhang's status to "pick-up completed," the order basic information acquisition unit advanced the order progress to "in progress," and the trip environment acquisition unit updated the current road conditions in real time: the main road traffic flow was stable, the elevated road entrance was open, and the destination was expected to be reached in 30 minutes. Mr. Zhang helped Mr. Wang put his belongings in the trunk, checked the vehicle's exterior and fuel level, got into the driver's seat, started the vehicle, and drove according to the elevated road route planned by the system. After driving for 15 minutes, when they were about to reach the elevated road exit, the trip environment acquisition unit suddenly captured the traffic warning signs and traffic police signals ahead, and simultaneously obtained the real-time information issued by the traffic department that "the elevated road exit is temporarily closed due to a traffic accident requiring scene clearing, and vehicles are prohibited from leaving." At the same time, the driver status acquisition unit showed that Mr. Zhang's vehicle had slowed down 500 meters before the control point, with the speed dropping from 80 km / h to 20 km / h, deviating from the original route trajectory. The real-time status monitoring unit of the anomaly warning module quickly compares this data with the anomaly judgment threshold of "critical road sections in the journey are impassable" to confirm the triggered route anomaly. The anomaly identification unit retrieves data from surrounding monitoring and traffic departments to analyze the cause of the anomaly as "temporary traffic accident leading to traffic control," ruling out road construction or weather effects. The level quantification unit considers that there is a surface expressway near the control point that can be used as an alternative route, and that the detour will only add 5 minutes to the journey and will not significantly affect the arrival time, so it sets the anomaly level as "minor" and generates a warning message: "Temporary traffic control at the elevated road exit ahead. A surface expressway detour route has been planned for you, with an estimated delay of 5 minutes." The strategy generation unit of the order linkage processing module combines real-time traffic data and uses algorithms to calculate the optimal detour route: exit the elevated road from the auxiliary road ahead, turn onto the surface expressway, pass through 2 traffic light intersections, and arrive at the community directly, while avoiding the construction section of the surface road. The platform linkage unit synchronizes the new route to Mr. Zhang's terminal in real time, and Mr. Zhang confirms the route adjustment through voice broadcast. The user interaction module then informed Mr. Wang of the route adjustment via voice prompt. Mr. Wang opened the visualization view on the APP, and a green detour route was added next to the original red route on the map. The "estimated arrival time" on the timeline was updated from 09:48 to 09:53, and the countdown timer for key nodes showed "15 minutes to the destination". He swiped the screen to view the traffic details of the detour route. After confirming that there was no congestion, he clicked the "Understand" button to complete the interaction, and the system recorded his feedback in the order data.

[0061] During the detour, the system continuously tracks and monitors the order status from multiple dimensions: the status acquisition module updates Mr. Zhang's real-time location and speed every 30 seconds, showing that the vehicle maintains a speed of around 60 km / h, which meets the speed limit requirements of the surface expressway. It also shows that the user's vehicle has 40% fuel remaining, eliminating the need for refueling. The trip environment acquisition unit monitors the traffic conditions of the surface road in real time, showing only a brief wait at one traffic light intersection, with the congestion level remaining at "low," and no new traffic control or accident information. The driver status acquisition unit shows that Mr. Zhang's terminal communication signal is stable, with no abnormal reports. The data integration unit of the multi-dimensional order tracking module deduplicates and standardizes this real-time data, generating a new order data snapshot, which, together with previous fault records and route adjustment records, forms a complete order lifecycle data chain. The spatiotemporal fusion tracking model dynamically adjusts the trajectory display in the visualization view based on the latest location data, ensuring that the trajectory Mr. Wang sees on the APP is completely synchronized with the actual driving route. The data dashboard updates information such as "remaining distance 8 km," "current road conditions good," and "estimated arrival time 09:53" in real time. The anomaly warning module is constantly monitoring, extracting anomaly-related data from the order tracking model every second and comparing it with preset thresholds. No new anomalies were detected, such as location stagnation, abnormal speed, or route deviation, and the previous "minor anomaly" status was automatically lifted. Mr. Wang wanted to confirm his arrival time during his journey and asked through the "Contact Customer Service" function on the app. Based on real-time order data synchronized with the system, customer service quickly replied: "Hello! The vehicle is currently traveling on a surface expressway, 8 kilometers from the destination. Road conditions are smooth, and we expect to arrive around 09:52." Mr. Wang expressed satisfaction with the reply, and the system recorded this customer service interaction in the user feedback database for future optimization of customer service response speed and data synchronization efficiency. During his drive, Mr. Zhang's terminal occasionally displayed a "No anomalies on the current route, please drive safely" prompt, reminding him to stay focused. It also displayed Mr. Wang's contact information and the vehicle's emergency contact information to ensure timely communication in case of emergencies.

[0062] At 9:52 PM, Mr. Zhang successfully arrived at the entrance of the "Yunxi Garden" residential complex. He parked the vehicle in the designated parking space, turned off the engine, and clicked the "Arrived at Destination" button on the terminal. The status collection module immediately updated the order progress to "Pending Confirmation of Completion," while simultaneously collecting the vehicle's final parking location and exterior condition, generating a final data snapshot before order completion. The multi-dimensional order tracking module integrated all order data to generate a complete order lifecycle tracking report. The report included all time points from Mr. Wang placing the order at 9:00 AM to his arrival at 9:52 AM, detailed records of two abnormal events (Mr. Li's electric vehicle malfunction, elevated highway exit control), corresponding handling solutions (changing the driver, detour), user feedback, and tamper-proof snapshot links for all status data. Mr. Zhang helped Mr. Wang open the car door, assisted him in retrieving items from the trunk, and after confirming the vehicle was undamaged, both parties completed the service confirmation on the terminal. The user interaction module sent Mr. Wang a notification that his order was pending confirmation. He checked the order details on the app, confirmed the service time, route, and cost were correct, clicked the "Confirm Completion" button, and paid the chauffeur fee. The system then sent an order summary: "This chauffeur order is complete! The total distance traveled was 33 kilometers, taking 52 minutes. Two issues were handled during the trip (both resolved satisfactorily). Driver Zhang's service rating is 4.9 out of 5. You can click the link to view the complete order tracking record and data snapshot. Feedback is welcome." Mr. Wang received a text message notification that the order was complete. He then opened the data snapshot review section in the app and viewed the location record when Mr. Li's malfunction occurred, Mr. Zhang's driving trajectory, and the time points of route adjustments, confirming that the entire process was transparent and traceable. After submitting his service summary report on the terminal, Mr. Zhang rode his electric bike away from the community. The system backend then started matching him with new nearby orders based on his location and status. Meanwhile, Mr. Wang carried his belongings into the community. Although he encountered two minor issues during this ride-hailing experience, the timely warnings and handling by the system prevented a major impact on his trip. He wrote "timely handling of issues and professional service" in his order review. These reviews will also be synchronized to the system, providing data support for optimizing subsequent driver dispatching and issue handling strategies.

[0063] In summary, this embodiment of the application, after a user submits an order, collects basic order information, driver status, and trip environment data in real time, generating a visual tracking view for the user to view. If the driver encounters an electric vehicle malfunction during the ride, the system quickly identifies a moderate anomaly, optimizes the algorithm to dispatch the nearest driver, and informs the user through multiple channels while receiving feedback. If temporary road closures occur during the trip, the system promptly detects minor anomalies, plans the optimal detour route, and synchronizes it with both the driver and the user. Throughout the process, relying on multi-dimensional data integration, precise anomaly warnings, coordinated handling, and efficient user interaction, the system effectively resolves two different scenarios of anomalies, ensuring transparency and traceability throughout the order lifecycle, ultimately delivering the user safely and on time to their destination, effectively improving the stability of the chauffeur service and user satisfaction.

[0064] Next, referring to the accompanying drawings, a method for a multi-dimensional status tracking and anomaly warning system for chauffeur service user orders is described according to an embodiment of this application.

[0065] like Figure 7 As shown, the method of this multi-dimensional status tracking and anomaly early warning system for designated driver user orders includes the following steps: In step S101, multi-dimensional status data of the designated driver order is obtained.

[0066] It is understandable that this application's embodiments, by acquiring multi-dimensional status data of chauffeur orders, lay a solid foundation for the entire process. This data covers key aspects such as basic order information, real-time driver status, and dynamic trip environment. It supports the generation of visual tracking views, allowing users to intuitively grasp order progress, and provides accurate evidence for anomaly identification, helping the system quickly detect abnormal signs such as vehicle malfunctions and sudden changes in road conditions. Simultaneously, comprehensive and real-time data accelerates the generation of optimal handling strategies, ensuring timely resolution of anomalies and improving service stability and user trust.

[0067] In step S102, the multi-dimensional status data of the ride-hailing order is integrated and processed, and an immutable data snapshot of the entire life cycle of the order is generated based on the timestamp. At the same time, a spatiotemporal fusion algorithm is used to construct a full life cycle tracking model of the order and generate a visual order status tracking view.

[0068] Among them, the tamper-proof data snapshot refers to the fixed recording of multi-dimensional status data of each key node in the entire life cycle of a ride-hailing order based on timestamps, generating an unmodifiable original data copy for subsequent traceability and verification.

[0069] It is understood that this application embodiment generates an immutable snapshot of the entire order lifecycle, fixing multi-dimensional status data at each key node to form an unmodifiable original copy. This provides reliable data support for building the order tracking model, while ensuring data traceability and verification. It not only assists in quickly tracing the cause and defining responsibility when anomalies occur, but also allows users to view complete data through the backtracking portal, enhancing their trust in the service. Furthermore, it provides accurate raw data basis for the platform to optimize algorithms and improve service quality.

[0070] In step S103, dynamic status data related to anomalies are extracted from the order lifecycle tracking model, compared with the anomaly judgment threshold, and the cause of the anomaly is analyzed by the feature matching algorithm. Order anomaly events are identified and the anomaly level is quantified to generate anomaly warning information.

[0071] Among them, the feature matching algorithm first extracts key features from the objects to be matched, and then establishes the feature correspondence between objects by calculating the similarity between features.

[0072] It is understood that this application embodiment uses a feature matching algorithm to compare the dynamic status data related to order anomalies with the anomaly judgment threshold. It accurately matches anomaly symptoms with preset multi-dimensional anomaly feature templates such as delays, location deviations, and driver anomalies, quickly analyzing the root cause of the anomaly and clarifying the specific anomaly type. This not only improves the accuracy and efficiency of anomaly identification, avoiding delays in handling due to ambiguous causes, but also provides crucial support for anomaly level quantification and targeted early warning information generation, making early warnings more valuable and facilitating the rapid initiation of corresponding handling strategies. This effectively enhances the anomaly response capability and user experience of the designated driver service.

[0073] For example, after Ms. Wang placed an order for a designated driver, the order entered the pick-up stage. The system extracted abnormal dynamic status data from the order's full lifecycle tracking model: the driver, Mr. Li, remained stationary at the intersection of Wensan Road and Jiaogong Road for 12 consecutive minutes, his driving speed remained at 0, the battery level of his registered electric vehicle plummeted from 58% to 7% with no charging record, and no "temporary parking" reporting was triggered. The feature matching algorithm then started, comparing these abnormal signs with the system's preset "equipment failure during the pick-up stage" abnormal feature template—this template clearly defines "location stationary for more than 8 minutes + speed of 0 + battery level below 15% + no reporting record" as the core feature combination of electric vehicle range failure. The algorithm calculated that the extracted abnormal data matched the template with a 96% accuracy rate, exceeding the preset matching threshold of 80%. The algorithm quickly analyzed the cause of the anomaly as a sudden battery drain failure of Mr. Li's electric vehicle, identifying it as an "abnormality of equipment failure during the pick-up phase." Considering that Ms. Wang had been waiting for 8 minutes and there were 2 backup drivers available in the nearby Wulin business district, the anomaly was quantified as a medium level. The system simultaneously generated a warning message: "The driver's electric vehicle has a battery drain failure. A backup driver has been dispatched nearby. The delay is expected to be 10 minutes." This provided a clear basis for subsequent dispatch strategies.

[0074] In step S104, based on the early warning information, the backend of the designated driver platform is contacted, and an improved multi-objective particle swarm optimization algorithm is used to generate the optimal order adjustment strategy and emergency response plan.

[0075] It is understood that this application embodiment, by combining early warning information with the backend of the designated driver platform and employing an improved multi-objective particle swarm optimization algorithm, efficiently generates an optimal order adjustment strategy and corresponding emergency response plan that balances minimizing delay time, maximizing user satisfaction, and minimizing scheduling costs. This achieves a seamless closed-loop connection from abnormal early warning to on-the-spot handling, and the algorithm optimization ensures the rationality and relevance of the plan. It can quickly coordinate platform resources to reduce the impact of service interruptions, effectively control scheduling costs, improve user experience, and comprehensively guarantee the stability and reliability of the designated driver service.

[0076] In step S105, the visualized order status tracking view, abnormal warning information, order adjustment strategy and emergency response plan are displayed on the user terminal, and a user feedback channel is provided.

[0077] The user feedback channel refers to the interactive entry point in the user-side interface that displays information related to the designated driver order, allowing users to express their opinions, make requests, or evaluate the service regarding the order status, abnormal situations, and handling solutions.

[0078] It is understood that this application embodiment, by setting up a user feedback channel, establishes a two-way interactive bridge between users and the platform, based on displaying a visual order status tracking view, abnormal warning information, order adjustment strategies, and emergency response plans on the user's end. This channel is specifically designed to receive user opinions, requests, or evaluations regarding the display effect of relevant information, the rationality of abnormal handling, and service quality. This not only protects users' right to express their opinions on order services and ensures that user requests are responded to in a timely manner, thereby improving service experience and trust, but also provides the platform with real service optimization basis, helping to iterate visualization functions, improve abnormal handling strategies, and further enhance the suitability of the chauffeur service and user satisfaction.

[0079] According to the embodiments of this application, a method for a multi-dimensional status tracking and anomaly warning system for ride-hailing user orders is proposed. This method comprehensively collects multi-dimensional data on basic order information, driver status, and travel environment through a status acquisition module. This breaks through the limitations of traditional single-dimensional data collection, providing complete and accurate basic data support for order tracking and anomaly identification, avoiding tracking deviations and missed warnings due to missing information. The multi-dimensional order tracking module performs deduplication, noise reduction, and standardization on the data. It constructs a full lifecycle tracking model based on a spatiotemporal graph convolutional network with an improved attention mechanism, generating a visual view containing real-time trajectories and timelines. This allows users to intuitively grasp the status of each stage of the order and supports snapshot backtracking, solving the problem of traditional tracking lacking spatiotemporal correlation and visualization. An anomaly warning module... By pre-setting multi-dimensional anomaly thresholds, the system monitors and captures anomaly signs in real time. It uses feature matching algorithms to analyze causes and quantify three levels of anomaly severity, enabling accurate identification and early warning of abnormal events, thus mitigating risks caused by delayed anomaly detection. The order linkage processing module, in conjunction with the platform backend, utilizes an improved multi-objective particle swarm optimization algorithm to generate optimal adjustment strategies and contingency plans, balancing delay time, user satisfaction, and scheduling costs, thereby improving the efficiency and rationality of anomaly handling. The user interaction module provides diverse feedback channels by pushing views, warnings, and solutions through multiple channels, ensuring users receive information promptly and participate in interactions. Synchronous feedback data supports dynamic strategy optimization, significantly improving the intelligence level of ride-hailing order management, anomaly response speed, and user experience, while reducing order risks and scheduling costs. This solves the problems of fragmented state information and delayed anomaly detection in existing technologies.

[0080] The following will illustrate a method for a multi-dimensional status tracking and anomaly warning system for ride-hailing user orders through a specific embodiment, such as... Figure 8 As shown, it includes: One evening at 9:00 PM, Mr. Liu, from the Chengdong Business District, finished a business engagement and placed an order through the "Changxing Daijia" APP, selecting the "Preferred Daijia" service—a service that specifies "drivers with over 3 years of driving experience, uniform attire, and equipped with shoe covers and seat covers." The system dispatched the order within 10 seconds, generating order number ORD20240068, and simultaneously notified that "Driver Mr. Zhao has accepted the order and is on his way on an electric scooter, expected to arrive at 9:08 PM."

[0081] The moment the order was generated, the platform initiated multi-dimensional data collection: on the one hand, it identified Mr. Liu's core needs—the departure point "South Gate of Maoye Building in the Business District", the destination "North Gate of Cultural Square in the City Center", and the expectation of arriving before 21:40; on the other hand, it used Mr. Zhao's driver app to capture the dynamic data of his electric mobility scooter in real time: current location (latitude and longitude approximately 113.8°E, 22.6°N), driving speed 23km / h, battery level 76%, and Mr. Zhao's real-time status (accepting orders, no violations, and a 98% positive feedback rate for nearly 30 orders). This data is integrated every 20 seconds and combined with timestamps to generate an immutable snapshot. For example, "Snapshot0068-210020" records "Mr. Zhao is 1.2 kilometers away from the pick-up point, speed is 23 km / h, and battery level is 75%". At the same time, a tracking model is built through a spatiotemporal fusion algorithm, and a visual view appears on Mr. Liu's APP interface: the green icon on the map represents Mr. Zhao's electric vehicle, which is moving towards him along the main road of the business district. The timeline below is marked "Order accepted successfully → 21:08 Expected pick-up → 21:35 Expected arrival". The dashboard on the right shows the battery level and speed of Mr. Zhao's electric vehicle in real time.

[0082] At 21:05, when Mr. Zhao reached the approach bridge of the cross-river bridge in the business district, his electric mobility scooter suffered a tire blowout. He stopped urgently in the emergency lane but did not promptly report the "temporary stop." The platform's monitoring system immediately detected abnormal data: Mr. Zhao's location remained unchanged for 60 seconds in the middle of the approach bridge (latitude and longitude 113.82°E, 22.59°N), his speed dropped to 0, and the scooter's battery level remained at 72% but showed no movement. The system automatically extracted this data and compared it with the "abnormal threshold for the pick-up phase" (location stationary for more than 40 seconds, speed of 0, no report triggered), triggering an early warning mechanism. Subsequently, the feature matching algorithm was activated, combining the features of "location stationary + speed of 0 + no report" with the preset "driver's mobility scooter malfunction" template (core features: location stationary for ≥40 seconds during pick-up + no report record), achieving a matching degree of 94%, far exceeding the 75% trigger threshold. The system immediately identified the "replacement vehicle malfunction" anomaly. Combining this with the scenario of "Mr. Liu did not wait beyond the allotted time and the vehicle was parked in a designated parking space," it quantified the anomaly as "moderate anomaly" and sent a warning to Mr. Liu: "Dear user, your order ORD20240068 has encountered an anomaly: Mr. Zhao's electric replacement vehicle has suffered a sudden tire blowout and is temporarily unable to continue his journey. We will immediately reassign you, and a solution will be provided within 1 minute." The message was sent simultaneously via SMS and an app pop-up.

[0083] Mr. Liu clicked the "Emergency Reassignment" button in the pop-up window—a core option in the platform's user feedback channel, which also includes "Inquiry Progress" and "Temporarily Waiting" options. Upon receiving the feedback, the platform's backend immediately invoked an improved multi-objective particle swarm optimization algorithm, using "delay time ≤ 15 minutes, driver satisfaction rate ≥ 97%, and dispatch distance ≤ 3 kilometers" as the objective function to calculate the available transportation capacity within a 5-kilometer radius. After 8 iterations, the algorithm identified the optimal solution: Mr. Sun, located 1.8 kilometers from Mr. Liu, had 5 years of driving experience, nearly 100 rides with a 99% satisfaction rate, and his electric scooter had 88% battery and was currently idle. An adjustment strategy was then generated: Mr. Sun would accept the order, arriving at the pick-up point at 21:12 and delivering to the destination at 21:42, a 7-minute delay from the original plan. A contingency plan was also finalized: Mr. Liu would receive a 10 yuan no-minimum-spend coupon, and Mr. Sun's real-time location, contact number, and a service commitment to "prioritize nighttime driving safety" would be pushed to him via the app.

[0084] This information was updated in real time in Mr. Liu's visualization: the green icon on the map switched to Mr. Sun's logo, the timeline updated to "Reassigning → 21:12 Expected Pickup → 21:42 Expected Arrival," the dashboard added a "Mr. Sun Service Rating 4.9 / 5" prompt, and the adjustment strategy and contingency plan were clearly listed below the warning information, retaining a feedback entry for "Disagreement with the plan." At 21:11, Mr. Sun arrived at the south gate of Maoye Building, showed his work ID, put seat covers on Mr. Liu's vehicle, confirmed the route, and set off. At 21:40, the vehicle safely arrived at the north gate of Cultural Square, Mr. Liu completed the payment through the APP, and the system automatically popped up the "Service Evaluation" entry.

[0085] After the service is completed, the platform generates a full-link data record, TraceORD20240068, which is permanently stored in the database. This record includes the order timeline (order placed at 21:00 → Mr. Zhao accepts the order at 21:01 → malfunction at 21:05 → reassignment to Mr. Sun at 21:06 → service completed at 21:40), snapshots of key nodes (such as "Snapshot0068-210500" recording Mr. Zhao's location and status at the time of the malfunction), algorithm iteration logs, and Mr. Liu's "emergency reassignment" operation record. This immutable data provides a basis for tracing potential service disputes and offers real-world case support for the platform to optimize the "replacement vehicle malfunction warning threshold" and "reassignment algorithm iteration parameters."

[0086] In summary, this application's embodiments, through end-to-end integration of multi-dimensional status data collection, tamper-proof snapshot generation, spatiotemporal fusion tracking model construction, feature matching anomaly identification, multi-objective algorithm optimization scheduling, and user feedback management, achieve real-time transparent monitoring of order status, intelligent and precise handling of abnormal events, and efficient collaborative allocation of service resources. This significantly improves the reliability and user experience of the chauffeur service, continuously optimizes anomaly handling efficiency through data-driven algorithm iteration, and strengthens user trust with tamper-proof end-to-end records, providing a replicable technological paradigm for the intelligent service upgrade of the chauffeur industry.

[0087] Figure 9 A schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may include: The memory 901, the processor 902, and the computer program stored on the memory 901 and capable of running on the processor 902.

[0088] When the processor 902 executes the program, it implements the method of a multi-dimensional status tracking and anomaly warning system for chauffeur user orders provided in the above embodiments.

[0089] Furthermore, electronic devices also include: Communication interface 903 is used for communication between memory 901 and processor 902.

[0090] The memory 901 is used to store computer programs that can run on the processor 902.

[0091] The memory 901 may include high-speed RAM (Random Access Memory) memory, and may also include non-volatile memory, such as at least one disk storage.

[0092] If the memory 901, processor 902, and communication interface 903 are implemented independently, then the communication interface 903, memory 901, and processor 902 can be interconnected via a bus to complete communication between them. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 9 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0093] Optionally, in a specific implementation, if the memory 901, processor 902, and communication interface 903 are integrated on a single chip, then the memory 901, processor 902, and communication interface 903 can communicate with each other through an internal interface.

[0094] The processor 902 may be a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of this application.

[0095] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described method for a multi-dimensional status tracking and anomaly warning system for chauffeur service user orders.

[0096] Furthermore, this application also provides a computer program product, including a computer program or instructions, which, when executed, implement the aforementioned method for a multi-dimensional status tracking and anomaly warning system for chauffeur service user orders.

[0097] In the description of this specification, the references to "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0098] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0099] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0100] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any of the following techniques known in the art, or a combination thereof: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0101] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0102] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A multi-dimensional status tracking and anomaly early warning system for chauffeur service user orders, characterized in that, include: The system includes a status acquisition module, a multi-dimensional order tracking module, an anomaly warning module, an order linkage processing module, and a user interaction module; among which, The status acquisition module is used to collect multi-dimensional status data of chauffeur orders in real time. The multi-dimensional order tracking module is used to integrate and process the multi-dimensional status data, construct an order lifecycle tracking model based on a spatiotemporal fusion algorithm, and generate a visual order status tracking view. The anomaly warning module is used to monitor changes in order status in real time, identify abnormal events in orders and quantify the level of abnormality by combining preset multi-dimensional anomaly judgment thresholds, and generate anomaly warning information. The order linkage processing module is used to receive the early warning information, link with the backend of the designated driver platform, and use the improved multi-objective particle swarm optimization algorithm to generate the optimal order adjustment strategy and emergency response plan. The user interaction module is used to push order status tracking views, abnormal warning information and handling solutions to the designated driver user's terminal, and provide a user feedback and interaction channel.

2. The multi-dimensional status tracking and anomaly early warning system for chauffeur service user orders according to claim 1, characterized in that, The status acquisition module includes an order basic information acquisition unit, a driver status acquisition unit, and a trip environment acquisition unit. The order basic information acquisition unit is used to collect static data such as order number, departure point, destination, and appointment time, and update the current order progress in real time. The driver status acquisition unit is used to synchronously collect the driver's real-time location, driving speed, and the order acceptance status, service status, and real-time communication status fed back by the driver's terminal. The trip environment acquisition unit is used to collect information such as congestion level of the trip route, weather conditions, road construction information, and traffic control information.

3. The multi-dimensional status tracking and anomaly early warning system for chauffeur service user orders according to claim 1, characterized in that, The multi-dimensional order tracking module includes a data integration unit, a spatiotemporal fusion tracking model construction unit, and a visualization view generation unit. The data integration unit performs deduplication, noise reduction, and format standardization on multi-dimensional status data to generate a unified dataset and generates an immutable snapshot of the entire order lifecycle based on timestamps. The spatiotemporal fusion tracking model construction unit uses a spatiotemporal graph convolutional network with an improved attention mechanism to process spatiotemporal data and construct an order lifecycle tracking model, mapping the order status in real time from creation, order acceptance, pick-up, trip, to completion. The visualization view generation unit generates a visual interface that includes a real-time trajectory map, a status timeline, key node countdowns, a multi-dimensional data dashboard, and a data snapshot backtracking entry, supporting user zooming and backtracking.

4. The multi-dimensional status tracking and anomaly early warning system for chauffeur service user orders according to claim 1, characterized in that, The anomaly warning module includes an anomaly rule storage unit, a real-time status monitoring unit, an anomaly identification unit, a level quantification unit, and an early warning generation unit. The anomaly rule storage unit is used to pre-store multi-dimensional anomaly judgment thresholds for time delays, location deviations, driver abnormalities, and resource shortages. The real-time status monitoring unit extracts dynamic status data related to anomalies from the order lifecycle tracking model every second, compares it with the multi-dimensional anomaly judgment thresholds, and captures anomaly symptoms. The anomaly identification unit, based on the anomaly symptoms, analyzes the cause of the anomaly using a feature matching algorithm to identify the specific anomaly type. The level quantification unit classifies anomalies into three levels: minor, moderate, and severe, and generates anomaly warning information containing the cause of the anomaly, the anomaly type, the scope of impact, and risk warnings.

5. The multi-dimensional status tracking and anomaly early warning system for chauffeur service user orders according to claim 1, characterized in that, The order linkage processing module includes an order anomaly handling strategy generation unit and a platform linkage unit. The order anomaly handling strategy generation unit receives the early warning information, extracts the anomaly type, level, and impact range, and generates order adjustment strategies for minor and moderate early warnings based on an improved multi-objective particle swarm optimization algorithm, with the objective function of "minimizing delay time, maximizing user satisfaction, and minimizing scheduling cost." For severe early warnings, it simultaneously generates order adjustment strategies and emergency response plans. The platform linkage unit links with the chauffeur platform backend through an API interface to synchronize order adjustment strategies, emergency response plans, execution status, and associated order data snapshots.

6. The multi-dimensional status tracking and anomaly early warning system for chauffeur service user orders according to claim 1, characterized in that, The user interaction module includes a view push unit, an early warning and solution push unit, and a feedback interaction unit. The view push unit displays the order status tracking view in real-time on the homepage and order details page of the chauffeur service app, and supports quick viewing via a floating window. The early warning and solution push unit uses a multi-channel push mode, including app pop-ups, SMS, and voice prompts, prioritizing the push of high-level early warning information and corresponding handling solutions. The feedback interaction unit provides interactive functions such as one-click confirmation of handling solutions, refusal of adjustments, consultation with customer service, and customized feedback. User feedback information is synchronized to the platform backend in real-time, supporting dynamic strategy optimization.

7. A method for applying to a multi-dimensional status tracking and anomaly warning system for chauffeur service user orders as described in any one of claims 1-6, characterized in that, The method includes: Obtain multi-dimensional status data for designated driver orders; The multi-dimensional status data of the designated driver order is integrated and processed, and an immutable data snapshot of the entire life cycle of the order is generated based on the timestamp. At the same time, a spatiotemporal fusion algorithm is used to construct a full life cycle tracking model of the order and generate a visual order status tracking view. Extract dynamic status data related to anomalies from the order lifecycle tracking model, compare it with the anomaly judgment threshold, analyze the cause of the anomaly through a feature matching algorithm, identify order anomaly events and quantify the anomaly level, and generate anomaly warning information; Based on the aforementioned warning information, the system will coordinate with the backend of the designated driver platform and utilize an improved multi-objective particle swarm optimization algorithm to generate the optimal order adjustment strategy and emergency response plan. The visualized order status tracking view, abnormal warning information, order adjustment strategies and emergency response plans are displayed on the user's end, and a user feedback channel is provided.

8. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor. The processor executes the program to implement the method of a multi-dimensional status tracking and anomaly warning system for chauffeur service user orders as described in claim 7.

9. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, When a computer program or instruction is executed, it implements the method of a multi-dimensional status tracking and anomaly warning system for chauffeur user orders as described in claim 7.

10. A computer program product, comprising a computer program or instructions, characterized in that, When a computer program or instruction is executed, it implements the method of a multi-dimensional status tracking and anomaly warning system for chauffeur user orders as described in claim 7.

Images