A full-cycle intelligent early warning method and system for ETC service anomalies

Abstract

The present invention provides a full-cycle intelligent early warning method and system for abnormal ETC services. The steps of the method include: obtaining in-vehicle unit information of a user based on the interaction between the in-vehicle unit of the user terminal and the roadside unit, where the in-vehicle unit information includes a user identifier and an account balance; determining a corresponding user account based on the user identifier and performing automatic deduction based on the account balance; after completing the automatic deduction, calling user account data based on the user identifier, where the account data includes account behavior data and account historical transactions; determining whether to trigger a renewal reminder determination based on the user's account balance, and during the processing of the renewal reminder determination, determining whether to give a renewal reminder based on the user historical transaction data. If a renewal reminder is given, the renewal reminder is added to the account behavior data; extracting user behavior data for a preset time period, performing predictive analysis, determining a user experience value through the predictive analysis, and determining whether to give an early warning to the user.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic toll collection systems, and particularly to a full-cycle intelligent early warning method and system for ETC service anomalies. Background Art

[0002] An electronic toll collection system (Electronic Toll Collection System, abbreviated as ETC), also known as a non-stop toll collection system, is an automated toll collection method implemented using advanced information and communication technologies. The system sets up antennas and vehicle type identification systems at the entrances and exits of toll roads, highways, bridges, tunnels, etc., and installs on-vehicle devices (such as ETC on-vehicle equipment or special electronic cards) on vehicles. When a vehicle passes through these toll stations, there is no need to stop and pay manually, and the system can automatically identify the vehicle information and deduct the corresponding toll from the pre-deposited fees.

[0003] Compared with manual toll collection, the electronic toll collection system has many significant advantages. First, in terms of traffic efficiency, the electronic toll collection system greatly improves the vehicle passing speed. Using the ETC system, vehicles do not need to slow down and stop when passing through toll stations. The toll collection time for each vehicle is usually less than two seconds, and the passing capacity of its toll collection lanes is 5 to 10 times that of manual toll collection lanes. This effectively alleviates traffic congestion in front of toll stations, reduces the waiting time for vehicles in line, and also reduces energy consumption and exhaust emissions generated by starting and stopping, which is beneficial to environmental protection. Second, the electronic toll collection system has obvious advantages in terms of toll collection accuracy. Since ETC toll collection is automatically completed by electronic devices, it avoids problems such as human operation errors or oversights that may occur in manual toll collection, thus ensuring the accuracy of toll collection. This not only reduces disputes caused by toll collection errors but also improves the transparency and fairness of toll collection.

[0004] However, the existing electronic toll collection systems are often designed only for toll collection and cannot manage the user's account. The existing technology lacks intelligent management of user accounts. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide a full-cycle intelligent early warning method for ETC service anomalies to eliminate or improve one or more defects existing in the prior art.

[0006] One aspect of the present invention provides a full-cycle intelligent early warning method for ETC service anomalies. The steps of this method include:

[0007] Obtaining the in-vehicle unit information of the user based on the interaction between the in-vehicle unit of the user terminal and the roadside unit, where the in-vehicle unit information includes user identification and account balance;

[0008] Determine the corresponding user account based on the user identification, and perform automatic deduction based on the account balance;

[0009] After completing the automatic deduction, call the user account data based on the user identification, where the account data includes account behavior data and account historical transactions;

[0010] Determine whether to trigger a renewal reminder determination based on the user's account balance. During the processing of the renewal reminder determination, determine whether to send a renewal reminder based on the user historical transaction data. If a renewal reminder is to be sent, add the renewal reminder to the account behavior data;

[0011] Extract the user behavior data for a preset time period, perform predictive analysis based on the user behavior data for the preset time period, determine the user experience value through predictive analysis, and determine whether to give an early warning to the user.

[0012] Adopting the above solution, this solution can first complete the automatic deduction for the user based on the interaction between the in-vehicle unit and the roadside unit. When the automatic deduction is completed, it can analyze the user's account data. First, determine whether to send a renewal reminder based on the account balance and the recent transactions of the account, and determine the user's experience through the user behavior data for a period of time, and can achieve early warning to ensure the user's experience.

[0013] In some embodiments of the present invention, in the step of determining whether to trigger a renewal reminder determination based on the user's account balance, compare the account balance with the renewal threshold. If the account balance is lower than the renewal threshold, trigger the renewal reminder determination.

[0014] In some embodiments of the present invention, in the step of determining whether to send a renewal reminder based on the user historical transaction data, count the number of transactions of the user in the historical transaction data for a second preset time period, compare the number of transactions with the threshold number of times. If the number of transactions is greater than the threshold number of times, send a renewal reminder.

[0015] In some embodiments of the present invention, the step of performing predictive analysis based on the user behavior data for a preset time period and determining the user experience value through predictive analysis includes:

[0016] Count the user behavior data in the preset time period, and determine the behavior number corresponding to each piece of data in the user behavior data based on a preset comparison table;

[0017] Combine the behavior numbers corresponding to the user behavior data in the preset time period to obtain a behavior vector;

[0018] Input the behavior vector into a pre-trained neural network model, and the neural network model outputs the user experience value.

[0019] In a specific implementation process, the pre-trained neural network model is a convolutional neural network model.

[0020] In some embodiments of the present invention, the data of the user behavior data includes primary behaviors and secondary behaviors, the look-up table includes a primary look-up table and a plurality of secondary look-up tables, the primary behaviors correspond one-to-one with the items in the primary look-up table, each of the secondary look-up tables corresponds to one item in the primary look-up table, and the items in the secondary look-up table correspond one-to-one with the secondary behaviors. In the step of determining the behavior number corresponding to each piece of data in the user behavior data based on the pre-set look-up table, the primary behavior of each piece of data in the user behavior data is determined based on the primary look-up table in the look-up table to determine the primary number, and the secondary behavior of each piece of data in the user behavior data is determined based on the secondary look-up table in the look-up table to determine the secondary number. The primary number and the secondary number of each piece of data in the user behavior data are combined to obtain the behavior number.

[0021] In some embodiments of the present invention, in the step of inputting the behavior vector into the pre-trained neural network model and the neural network model outputs the user experience value, the neural network model inputs the user experience vector, and the value of each dimension in the user experience vector is the user experience value of an item of a user's sense of experience.

[0022] In some embodiments of the present invention, in the step of determining the user experience value through predictive analysis and determining whether to give an early warning to the user, the user experience value of each dimension in the user experience vector is compared with the corresponding experience threshold. If the user experience value of any dimension is lower than the corresponding experience threshold, it is determined that an early warning for the item of this sense of experience is triggered.

[0023] In some embodiments of the present invention, the steps of the method further include:

[0024] Compare the data in the in-vehicle unit information with the data in the server, and determine whether the data of the two corresponds. If any one of the data of the two does not correspond, an alarm reminder is given.

[0025] In some embodiments of the present invention, in the step of comparing the data in the in-vehicle unit information with the data in the server, determining whether the data of the two corresponds, and if any one of the data of the two does not correspond, giving an alarm reminder, the deduction information for a corresponding transaction is extracted. If there is no deduction information for a transaction, it is determined that there is a deduction omission; if there is a one-to-one corresponding deduction information for a transaction, it is determined that the transaction is normal; if there are multiple deduction information for a transaction, it is determined that there is a duplicate deduction, and an alarm reminder is given for the cases of deduction omission and duplicate deduction.

[0026] The second aspect of the present invention further provides a full-cycle intelligent early warning system for abnormal ETC services. The system includes a computer device, the computer device includes a processor and a memory, computer instructions are stored in the memory, and the processor is used to execute the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps implemented by the method described above.

[0027] The third aspect of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps implemented by the aforementioned full-cycle intelligent early warning method for abnormal ETC services.

[0028] The additional advantages, objectives, and features of the present invention will be partially described below, and will become partially apparent to those of ordinary skill in the art after studying the following text, or can be learned from the practice of the present invention. The objectives and other advantages of the present invention can be pointed out and obtained specifically in the description and the accompanying drawings.

[0029] Those skilled in the art will understand that the objectives and advantages that can be achieved by the present invention are not limited to the above specifically described, and the above and other objectives that the present invention can achieve will be more clearly understood according to the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The drawings described herein are used to provide a further understanding of the present invention, form a part of this application, and do not limit the present invention.

[0031] Figure 1 It is a schematic diagram of an embodiment of the full-cycle intelligent early warning method for abnormal ETC services of the present invention;

[0032] Figure 2 It is a schematic diagram of the overall framework of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0033] To make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below in conjunction with the embodiments and the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but do not limit the present invention.

[0034] Here, it should also be noted that in order to avoid obscuring the present invention due to unnecessary details, only the structures and / or processing steps closely related to the solution of the present invention are shown in the drawings, and other details less related to the present invention are omitted.

[0035] The disadvantages of the prior art include:

[0036] 1. Manual monitoring system: Currently, many ETC services rely heavily on manual monitoring and intervention. This method is inefficient and prone to missing problems, unable to achieve real-time monitoring and rapid response;

[0037] 2. Simple automation tools: Some automation tools can handle specific types of exceptions, such as automatic refunds or blacklist management, but these tools usually lack comprehensive analysis and prediction capabilities;

[0038] 3. Static data analysis: Some systems use regular data analysis to identify exceptions, but this method cannot capture and respond to the latest transaction exceptions in real time;

[0039] 4. Single-function systems: Existing systems often are designed for a single problem, such as a separate refund processing system or blacklist management system, lacking integration and unable to comprehensively solve various exceptions in ETC services;

[0040] 5. Customer service system: The customer service system usually intervenes after user complaints to provide solutions, but this method is passive and has a long response time.

[0041] 6. Lack of real-time performance: Existing systems often cannot achieve real-time monitoring and rapid response to ETC service exceptions, resulting in delays in problem discovery and resolution.

[0042] 7. Lack of predictability: Many existing solutions lack the ability to use data analysis to predict future potential exceptions and mainly rely on passive responses after problems occur.

[0043] 8. Low processing efficiency: Existing exception handling processes usually involve multiple departments and manual operations, resulting in low processing efficiency and long user waiting times.

[0044] 9. Insufficient personalized service: Existing systems often adopt a general processing method and lack personalized services for different user characteristics.

[0045] 10. Low degree of automation: After exception identification, existing systems often require manual intervention and lack an automated processing flow, which limits the processing speed and consistency.

[0046] 11. Poor user experience: Due to the above limitations, users often have a poor experience when encountering ETC service exceptions and need to wait a long time for problem resolution.

[0047] The advantages of this solution include:

[0048] 1. Improve real-time performance: The present invention uses real-time monitoring technology to quickly identify exceptions in ETC services, thereby achieving instant response.

[0049] 2. Enhanced prediction ability: By leveraging advanced data analysis and machine learning algorithms, the present invention can predict potential service anomalies and achieve proactive warning.

[0050] 3. Improved processing efficiency: Through an automated anomaly handling process, the present invention can reduce manual intervention and improve processing speed and efficiency.

[0051] 4. Personalized service implementation: The present invention can provide customized warning and handling solutions based on users' behavior patterns and historical data, enhancing the user experience.

[0052] 5. Automated processing: The automated processing mechanism of the present invention aims to reduce manual operations, ensure consistency, and reduce errors.

[0053] 6. User experience improvement: Through fast, personalized, and automated services, the present invention aims to improve user satisfaction and loyalty.

[0054] As Figure 1 and 2 shown, the present invention proposes a full-cycle intelligent warning method for ETC service anomalies, and the steps of this method include:

[0055] Step S100, obtaining the in-vehicle unit information of the user based on the interaction between the in-vehicle unit of the user side and the roadside unit, where the in-vehicle unit information includes user identification and account balance;

[0056] In the specific implementation process, OBU (On-Board Unit): the in-vehicle unit, is an ETC device installed on the vehicle and used to communicate with the roadside unit (RSU) to achieve automatic toll deduction; RSU (Road-Side Unit): the roadside unit, is a device installed at the toll station and communicates with the OBU to process transaction information.

[0057] In the specific implementation process, there is a bound card signature information between the ETC user account and the OBU, and the card signature information includes the card number and signature number, which are used to identify and verify the user identity.

[0058] Specifically, in the ETC system, due to reasons such as arrears, violations, or others, some ETC accounts will be marked as in the blacklist status and prohibited from using ETC services.

[0059] Step S200, determining the corresponding user account based on the user identification and performing automatic toll deduction based on the account balance;

[0060] In the specific implementation process, in the step of automatically deducting fees based on the account balance, deduct the fees from the corresponding bank card account based on the account balance, and compare whether the balance of the bank card account is consistent with the account balance in the in-vehicle unit information before deducting the fees. If they are inconsistent, record the discrepancy between the card and the account; if they are consistent, perform automatic fee deduction.

[0061] Step S300, after completing the automatic fee deduction, call the user account data based on the user identifier, and the account data includes account behavior data and account historical transactions;

[0062] Step S400, determine whether to trigger the renewal reminder determination based on the user's account balance. In the process of the renewal reminder determination, determine whether to send a renewal reminder based on the user historical transaction data. If a renewal reminder is to be sent, add the renewal reminder to the account behavior data;

[0063] Step S500, extract the user behavior data for a preset time period, perform predictive analysis based on the user behavior data for the preset time period, determine the user experience value through the predictive analysis, and determine whether to give an early warning to the user.

[0064] In the specific implementation process, this solution covers the entire service cycle from the user registering for the ETC account, to transaction processing during use, to handling possible abnormal situations, and until the user cancels the account.

[0065] The steps of this solution also include data integration and cleaning: by integrating the source data from different business systems, including the complaint system, the ETC business handling system, the card and label system, and the recharge and consumption refund system, this system can provide a unified data view. The data cleaning and processing module is responsible for ensuring the accuracy and consistency of the data, providing a high-quality data basis for subsequent analysis.

[0066] Adopting the above solution, this solution can first complete the automatic fee deduction for the user based on the interaction between the in-vehicle unit and the roadside unit. When the automatic fee deduction is completed, it can analyze the user's account data. First, determine whether to send a renewal reminder based on the account balance and the recent transactions of the account, and determine the user experience through the user behavior data for a period of time, and can achieve early warning to ensure the user experience.

[0067] In some embodiments of the present invention, in the step of determining whether to trigger the renewal reminder determination based on the user's account balance, compare the account balance with the renewal threshold. If the account balance is lower than the renewal threshold, trigger the renewal reminder determination.

[0068] In some embodiments of the present invention, in the step of determining whether to give a renewal reminder based on the user's historical transaction data, the number of transactions of the user in the historical transaction data for a second preset time period is counted, and the number of transactions is compared with a threshold number. If the number of transactions is greater than the threshold number, a renewal reminder is given.

[0069] In some embodiments of the present invention, the step of performing predictive analysis on the user behavior data based on a preset time period and determining the user experience value through predictive analysis includes:

[0070] Count the user behavior data in a preset time period, and determine the behavior number corresponding to each piece of data in the user behavior data based on a preset comparison table;

[0071] Combine the behavior numbers corresponding to the user behavior data in the preset time period to obtain a behavior vector;

[0072] Input the behavior vector into a pre-trained neural network model, and the neural network model outputs the user experience value.

[0073] In some embodiments of the present invention, the data acquisition module serves as the data input layer of the entire system, mainly responsible for real-time data acquisition and unified management in a multi-source heterogeneous environment, and supporting the efficient operation and data governance requirements of the big data system. The module covers the functions of extracting, transmitting, and preprocessing the original data of multiple business systems, laying a foundation for subsequent data governance, analysis, and application. These data include customer complaint records, ETC transaction records, card signature status, and account balances, etc.

[0074] Specifically, the data acquisition module architecture includes the following functional components:

[0075] a. Multi-source data interface manager:

[0076] Provide a standardized interface to support obtaining data from different business systems (such as the 12345 complaint system, ETC processing system).

[0077] Support multiple data access methods such as RESTful API, JDBC, and file system.

[0078] Automatically adapt to the connection and configuration of heterogeneous data sources according to the characteristics of different business systems.

[0079] b. Data acquisition engine:

[0080] Implement data acquisition based on the DataX platform, and support parallel extraction and loading of multi-source data.

[0081] Provide flexible data conversion functions to standardize the data into a unified format (such as JSON or table form).

[0082] c. Data cleaning module:

[0083] Automatically clean the collected data, including removing redundant data, completing fields, converting data types, etc.

[0084] Use a rule engine to perform quality verification on the data to ensure that the data meets predefined quality standards.

[0085] d. Data synchronization mechanism:

[0086] Based on incremental synchronization technology, support the efficient collection and transmission of real-time updated data.

[0087] Achieve real-time data synchronization by comparing the change logs of the source system (such as MySQL Binlog or Kafka).

[0088] e. Scheduling and governance module:

[0089] Use Dolphin Scheduler to achieve automatic scheduling of collection tasks.

[0090] Support task dependency management, multi-task parallel processing, and failure retry mechanism to ensure the high availability of data collection.

[0091] The collection process of the data collection module includes:

[0092] a. Access data sources:

[0093] Configure multi-source data interfaces according to the characteristics of different business systems.

[0094] Define the mapping relationship (field names, types, etc.) between the source system and the target storage.

[0095] b. Data collection:

[0096] Start a DataX task to collect raw data and load it into storage (such as HDFS).

[0097] c. Data cleaning and transformation:

[0098] After the data collection is completed, clean the data through a rule engine.

[0099] Convert the data format (such as JSON to table).

[0100] d. Data synchronization and governance:

[0101] Achieve real-time data synchronization based on Dolphin Scheduler.

[0102] Generate log records for the collected data to facilitate auditing and tracking.

[0103] Specifically, it mainly analyzes various abnormal behaviors that users may complain about, including 12345 complaints, unresolved orders and complaints, abnormal account closure and refund, etc. It is scheduled regularly through the big data processing platform every day. The Spark framework extracts all data from the DWD layer and ADS layer of the Hive data warehouse to cover the ADS layer. Then, the big data processing platform continues to schedule, and incrementally writes data from the Hive mapping table in the ClickHouse library to the ClickHouse result table in an ETL manner. After that, a small file merging operation is performed, and data other than the analysis of the current day is deleted from the ClickHouse result table.

[0104] Real-time data analysis: Real-time guessing of your complaints uses pseudo-real-time scheduling analysis. It analyzes the business data written from the materialized view to the real-time table in ClickHouse, and writes the results to the ClickHouse result table after the analysis is completed.

[0105] Technical implementation:

[0106] 1. Spark framework scheduled scheduling:

[0107] Use the Spark framework for scheduled scheduling to extract data from the DWD (Data Workspace Layer) and ADS (Analytical Data Service Layer) of the Hive data warehouse.

[0108] The DWD layer stores the data after data cleaning, and the ADS layer is the data suitable for analysis after processing and summarization.

[0109] Extract all data from these dimension tables and fact tables through Spark to ensure coverage of all data in the Hive ADS layer and complete data synchronization.

[0110] 2. Data ETL processing:

[0111] Perform ETL (Extract, Transform, Load) processing through the big data processing platform, and incrementally write the data of the Hive mapping table into ClickHouse.

[0112] 3. Data incremental writing:

[0113] Extract data from the ClickHouse mapping table and incrementally write it into the ClickHouse result table to ensure the currency of historical data, reduce the repeated writing of all data, and improve efficiency.

[0114] 4. Small file merging operation

[0115] A large number of small files may be generated during the incremental writing process. The small file merging operation can optimize disk storage and query performance.

[0116] ClickHouse supports efficient distributed queries, and the merge operation can significantly improve query efficiency.

[0117] 5. Data deletion operation

[0118] Delete the data in the ClickHouse result table that is not from the current day to avoid the accumulation of useless data, save storage space, and optimize query efficiency.

[0119] 6. Near-real-time scheduling

[0120] Adopt a near-real-time scheduling scheme. By processing in small batches at regular intervals (such as scheduling every minute), reduce data latency and resource consumption, and achieve near-real-time data updates.

[0121] 7. Efficient query

[0122] The distributed query architecture and columnar storage method of ClickHouse make the query based on user numbers very efficient.

[0123] The data is stored by column, and the data can be read according to the user number, which improves the page response speed.

[0124] In some embodiments of the present invention, the data of the user behavior data includes primary behaviors and secondary behaviors, the comparison table includes a primary comparison table and multiple secondary comparison tables, the primary behaviors correspond one-to-one with the items in the primary comparison table, each secondary comparison table corresponds to one item in the primary comparison table, and the items in the secondary comparison table correspond one-to-one with the secondary behaviors. In the step of determining the behavior number corresponding to each data in the user behavior data based on the preset comparison table, determine the primary behavior of each data in the user behavior data based on the primary comparison table in the comparison table to determine the primary number, determine the secondary behavior of each data in the user behavior data based on the secondary comparison table in the comparison table to determine the secondary number, and combine the primary number and the secondary number of each data in the user behavior data to obtain the behavior number.

[0125] In some embodiments of the present invention, the primary behaviors include incorrect deduction, device failure, and poor customer service experience; the secondary behaviors corresponding to incorrect deduction include system mis-deduction, repeated deduction, and amount discrepancy; the secondary behaviors corresponding to device failure include in-vehicle OBU malfunction and unrecognizability; the secondary behaviors corresponding to poor customer service experience include difficult-to-connect hotline, templated response, and poor attitude.

[0126] In the specific implementation process, the user behavior data includes repeated deductions, abnormal refunds, and approaching expiration of card signatures. This solution utilizes key element data, and the intelligent algorithm analysis module can identify and predict potential business anomalies, such as repeated deductions, abnormal refunds, approaching expiration of card signatures, etc. This prediction mechanism enables the system to take proactive measures before problems occur, thereby improving customer satisfaction and business processing efficiency.

[0127] Specifically, this solution also includes a user reach module. The user reach module is designed to communicate effectively with customers, promptly notify them about the status of ETC cards and OBU tags, as well as any abnormal transactions that require their attention. At the same time, this module is closely linked to the business system to ensure that once an anomaly is detected, it can respond quickly and take actions. Further, this solution can automate the processing flow: by designing automated business processing flows, such as automatically initiating refunds, card signature extensions, recharge supplements, etc., it reduces manual intervention and improves the processing speed and accuracy.

[0128] The user reach module serves as the hub of the entire intelligent early warning and processing system for ETC service anomalies, establishing a connection between the intelligent analysis module and the actual business execution system. When the intelligent analysis module accurately determines the anomalies existing in the ETC service based on data and algorithms, the business system reach module converts these early warning messages into practical specific business operation instructions. It accurately identifies the business processes corresponding to different types of abnormal early warnings to ensure that these abnormal early warning messages are delivered to the backend business execution system, thereby achieving timely processing of ETC service anomalies and ensuring the stable operation of the entire ETC service system.

[0129] Technical implementation:

[0130] 1. Design a business rule engine:

[0131] Build a business rule library containing various ETC service anomaly handling rules, such as the overdue warning rule (when the user account balance is lower than the threshold and the continuous usage times reach a certain amount, an automatic renewal reminder is triggered, which can be pushed via text messages and in-app notifications).

[0132] Use rule engine technology to achieve dynamic configuration management of rules, providing a visual editing interface to facilitate business personnel to modify rules according to requirements. For example, when the ETC service policy changes, the rules related to the overdue threshold can be quickly updated.

[0133] The rule engine matches rules based on real-time early warning information and executes operations. For example, when receiving a warning that the card signature is about to expire, it triggers a reminder process and selects the notification method according to the user's habits.

[0134] 2. Implement the API interface with the backend business system:

[0135] Design a standardized and secure API interface specification, adopting the RESTful architectural style to ensure smooth communication with the backend ETC-related business systems.

[0136] Develop API client libraries for different business systems, encapsulate the interaction details, and provide a simple calling method. For example, when interacting with the account management system, the client library provides function methods such as querying balance, updating status, and renewing fees.

[0137] Pay attention to the security and integrity of API interface data transmission, adopt the HTTPS protocol for encryption, conduct identity authentication and authorization management to ensure system security.

[0138] 3. Design a manual intervention mechanism:

[0139] Build a manual intervention operation platform for reviewers to view the detailed information of abnormal warnings and relevant business data. The platform provides an intuitive interface to display key information, historical data, and processing suggestions to assist in decision-making.

[0140] Establish a manual intervention process to clarify the steps and division of responsibilities for reviewers to handle abnormal warnings. Reviewers verify evidence according to the process and record the processing process and results to ensure traceable auditing.

[0141] Achieve seamless connection between manual intervention and automated processing processes. The results of manual processing update business data, feedback to the intelligent analysis module, and at the same time mark and record the processing situation for statistical analysis to improve the mechanism.

[0142] In some embodiments of the present invention, in the step of inputting the behavior vector into the pre-trained neural network model and the neural network model outputting the user experience value, the neural network model inputs the user experience vector, and the value of each dimension in the user experience vector is the user experience value of an item of a user's experience feeling.

[0143] In some embodiments of the present invention, in the step of determining the user experience value through predictive analysis and determining whether to give an early warning to the user, compare the user experience value of each dimension in the user experience vector with the corresponding experience threshold. If the user experience value of any dimension is lower than the corresponding experience threshold, it is determined to trigger an early warning for the item of this experience feeling.

[0144] In the specific implementation process, the user reach module is responsible for actively communicating with the user when detecting an abnormality in the ETC service during the ETC service exception handling, providing the details of the abnormality and the solution. Whether it is an ETC card, an OBU tag, or a transaction abnormality, the user can be informed in a timely manner to enhance user trust and satisfaction. Considering the user scenario and convenience, it can be pushed to the user through methods such as APP, SMS, or email.

[0145] Technical implementation:

[0146] 1. Integrated communication interface:

[0147] Develop a communication interface framework to support SMS, email, and application notifications (WeChat official account messages, mobile APP push). For SMS, cooperate with providers to access the gateway, accurately set parameters, and monitor the status. Configure the server connection for email, be compatible with common clients, and set various formats of content. Develop application notifications for different platforms, use relevant technologies for accurate push, and customize content to guide user operations.

[0148] 2. Design the user interaction process:

[0149] Draw a detailed interaction flow chart covering the operation process after the user receives a warning. The interaction design follows the principles of habit and convenience. For example, clicking on the notification jumps to the details page to display key information and operation buttons, adopts a responsive design to ensure the multi-device experience, provides real-time operation feedback, and records and analyzes behaviors to optimize the process.

[0150] 3. Implement personalized notification strategies:

[0151] Build a user portrait database, integrate multi-dimensional information to analyze user needs and preferences. Based on the analysis results, formulate strategy rules, such as selecting the notification time according to travel habits, providing detailed expense explanations for cost-sensitive users, and customizing the channel combination according to the user's notification method preferences.

[0152] With the rapid development of intelligent transportation systems, ETC services, as an important part of intelligent transportation management, play a key role in improving road use efficiency and reducing operating costs. However, the complexity of the ETC system and the growth of the user base also bring a series of challenges, especially in the monitoring, warning, and handling of service anomalies. Service anomalies not only affect the user experience but may also lead to economic losses and management difficulties. This solution analyzes the abnormal data in the user portrait to predict the items with poor user experience, that is, the possible complaint points of users, so that customer service can prepare in advance or actively solve problems.

[0153] In some embodiments of the present invention, the steps of the method further include:

[0154] Compare the data in the on-vehicle unit information with the data in the server, determine whether the data of the two corresponds, and if any one of the data of the two does not correspond, give an alarm reminder.

[0155] Adopting the above solution, the system realizes real-time monitoring of ETC transactions and account status, can timely capture problems such as unuploaded transaction information and inconsistent card account funds, and provides a feedback mechanism for quick adjustment and optimization of services.

[0156] In some embodiments of the present invention, in the step of comparing the data in the vehicle unit information with the data in the server to determine whether the data of the two corresponds, and if any one of the data of the two does not correspond, an alarm reminder is given. For a corresponding transaction, the deduction information is extracted. If there is no deduction information for a transaction, it is determined that there is a deduction omission; if there is one-to-one corresponding deduction information for a transaction, it is determined that the transaction is normal; if there are multiple deduction information for a transaction, it is determined that there is duplicate deduction, and an alarm reminder is given for the cases of deduction omission and duplicate deduction.

[0157] An embodiment of the present invention further provides a full-cycle intelligent early warning system for ETC service anomalies. The system includes a computer device, the computer device includes a processor and a memory, the memory stores computer instructions, and the processor is used to execute the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps implemented by the method described above.

[0158] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps implemented by the aforementioned full-cycle intelligent early warning method for ETC service anomalies. The computer-readable storage medium may be a tangible storage medium, such as a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a floppy disk, a hard disk, a removable storage disk, a CD-ROM, or any other form of storage medium known in the technical field.

[0159] Those of ordinary skill in the art should understand that the various exemplary components, systems, and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software, or a combination of both. Specifically, whether to implement in hardware or software depends on the specific application and design constraints of the technical solution. A professional technician can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present invention. When implemented in hardware, it can be, for example, an electronic circuit, an application-specific integrated circuit (ASIC), appropriate firmware, a plug-in, a functional card, etc. When implemented in software, the elements of the present invention are programs or code segments used to perform the required tasks. The program or code segment can be stored in a machine-readable medium or transmitted through a data signal carried in a carrier wave on a transmission medium or a communication link.

[0160] It should be clear that the present invention is not limited to the specific configurations and processes described above and illustrated in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and illustrated as examples. However, the method process of the present invention is not limited to the specific steps described and illustrated, and those skilled in the art can make various changes, modifications, and additions, or change the order between steps after understanding the spirit of the present invention.

[0161] In the present invention, features described and / or illustrated for one embodiment can be used in the same or a similar manner in one or more other embodiments, and / or combined with the features of other embodiments or replace the features of other embodiments.

[0162] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, various changes and modifications can be made to the embodiments of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A full-cycle intelligent early warning method for abnormal ETC services, characterized in that, The steps of the method include: Obtaining the in-vehicle unit information of the user based on the interaction between the in-vehicle unit of the user terminal and the roadside unit, where the in-vehicle unit information includes user identification and account balance; Determining the corresponding user account based on the user identification and automatically deducting fees based on the account balance; After completing the automatic fee deduction, calling the user account data based on the user identification, where the account data includes account behavior data and account historical transactions; Determining whether to trigger a renewal reminder determination based on the user's account balance. During the processing of the renewal reminder determination, determining whether to send a renewal reminder based on the user's historical transaction data. If a renewal reminder is to be sent, adding the renewal reminder to the account behavior data; Extracting the user behavior data for a preset time period, performing predictive analysis based on the user behavior data for the preset time period, determining the user experience value through predictive analysis, counting the user behavior data in the preset time period, determining the behavior number corresponding to each piece of data in the user behavior data based on a preset comparison table; combining the behavior numbers corresponding to the user behavior data in the preset time period to obtain a behavior vector; inputting the behavior vector into a pre-trained neural network model, where the neural network model outputs the user experience value and determines whether to give an early warning to the user.

2. The full-cycle intelligent early warning method for ETC service anomalies according to claim 1, wherein In the step of determining whether to trigger a renewal reminder determination based on the user's account balance, comparing the account balance with the renewal threshold. If the account balance is lower than the renewal threshold, triggering a renewal reminder determination.

3. The full-cycle intelligent early warning method for ETC service anomalies according to claim 2, characterized in that In the step of determining whether to send a renewal reminder based on the user's historical transaction data, counting the number of transactions of the user in the historical transaction data for a second preset time period, comparing the number of transactions with the threshold number of times. If the number of transactions is greater than the threshold number of times, sending a renewal reminder.

4. The full-cycle intelligent early warning method for ETC service anomalies according to claim 1, wherein The data of the user behavior data includes primary behaviors and secondary behaviors, and the comparison table includes a primary comparison table and multiple secondary comparison tables. The primary behaviors correspond one-to-one with the items in the primary comparison table. Each secondary comparison table corresponds to one item in the primary comparison table, and the items in the secondary comparison table correspond one-to-one with the secondary behaviors. In the step of determining the behavior number corresponding to each piece of data in the user behavior data based on the preset comparison table, determining the primary behavior of each piece of data in the user behavior data based on the primary comparison table in the comparison table to determine the primary number, determining the secondary behavior of each piece of data in the user behavior data based on the secondary comparison table in the comparison table to determine the secondary number, and combining the primary number and the secondary number of each piece of data in the user behavior data to obtain the behavior number.

5. The full-cycle intelligent early warning method for ETC service anomalies according to claim 4, wherein, In the step of inputting the behavior vector into a pre-trained neural network model and the neural network model outputting the user experience value, the neural network model inputs a user experience vector, and the value of each dimension in the user experience vector is the user experience value of an item of a user's sense of experience.

6. The full-cycle intelligent early warning method for ETC service anomalies according to claim 5, characterized in that, In the step of determining the user experience value through predictive analysis and determining whether to give an early warning to the user, the user experience value of each dimension in the user experience vector is compared with the corresponding experience threshold. If the user experience value of any dimension is lower than the corresponding experience threshold, it is determined to trigger an early warning for the item that triggers this kind of experience feeling.

7. The full-cycle intelligent early warning method for ETC service anomalies according to claim 1, wherein The steps of the method further include: Comparing the data in the in-vehicle unit information with the data in the server, determining whether the data of the two correspond. If any one of the data of the two does not correspond, an alarm reminder is given.

8. The full-cycle intelligent early warning method for ETC service anomalies according to claim 7, wherein In the step of comparing the data in the in-vehicle unit information with the data in the server, determining whether the data of the two correspond. If any one of the data of the two does not correspond, an alarm reminder is given. For a corresponding transaction, the deduction information is extracted. If there is no deduction information for a transaction, it is determined that there is a deduction omission; if there is a one-to-one corresponding deduction information for a transaction, it is determined that the transaction is normal; If there are multiple deduction information for a transaction, it is determined that there is a duplicate deduction, and an alarm reminder is given for the cases of deduction omission and duplicate deduction.

9. A full-cycle intelligent early warning system for ETC service anomalies, characterized in that, The system includes a computer device, the computer device includes a processor and a memory, the memory stores computer instructions, and the processor is used to execute the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps implemented by the method according to any one of claims 1 to 8.

Images