Road traffic intelligent operation method and system based on multi-source data fusion

Abstract

The present application relates to the technical field of data analysis, in particular to a multi-source data fusion road traffic intelligent operation and maintenance method and system, in the present application, a three-dimensional model is constructed by a credibility calculation module by comprehensively considering three types of factors of road environment, device state and historical verification, a dynamically updated credibility index is generated, and the weight is corrected in real time based on feedback, solving the problem that the reliability of multi-source data dynamically fluctuates due to environmental interference, device aging and historical deviation; a road feature conflict resolution module maps the dynamic credibility index to the feature dimension corresponding to the road topology conflict feature, constructs a road conflict decision matrix, and generates a road anomaly decision set with a credibility label according to the feature dimension fusion; a resource scheduling module divides three response zones based on the credibility index, combines the device priority and the credibility label to perform precise scheduling with three constraints and generates a resource-decision correlation graph, realizing adaptive optimization configuration of limited maintenance resources according to data reliability.

Description

Technical Field

[0001] This invention relates to the field of data analysis technology, specifically to a method and system for intelligent operation and maintenance of road traffic involving multi-source data fusion. Background Technology

[0002] Existing road traffic operation and maintenance systems heavily rely on heterogeneous multi-source data (such as road network monitoring sensor data, environmental sensor data, historical maintenance records, dynamic traffic flow data, etc.), but face significant technical challenges in practical applications:

[0003] Different data sources have varying degrees of reliability and exhibit dynamic fluctuations due to differences in acquisition accuracy, fluctuations in the operating status of the equipment itself, external environmental interference (such as the impact of severe weather on sensors), and insufficient completeness of historical verification information. This directly weakens the reliable foundation of decision-making.

[0004] Meanwhile, conflicts frequently occur at the feature level among multi-source data, such as inconsistencies in the state judgments of the same road segment by different devices (geometric or topological connection contradictions), and logical discrepancies between historical data patterns and real-time observation data. Existing methods struggle to efficiently and objectively identify and integrate these conflicting features, leading to confusion in decision-making.

[0005] In addition, traditional resource scheduling mechanisms are mostly based on static rules or preset thresholds, which cannot effectively respond to real-time changes in the credibility of multi-source data and the results of conflict resolution. This leads to a lack of precision and dynamic adaptability in resource allocation strategies, often manifesting as delayed response or resource mismatch.

[0006] Therefore, there is an urgent need for a smart operation and maintenance system that can integrate multi-source data, dynamically quantify and evaluate data credibility, automatically resolve feature conflicts, and thereby achieve intelligent hierarchical scheduling of resources. Summary of the Invention

[0007] The purpose of this invention is to provide a road traffic intelligent operation and maintenance method and system based on multi-source data fusion, in order to solve the problems mentioned in the background art. Specific technical problems include:

[0008] How to solve the problem of dynamic fluctuations in the reliability of multi-source data caused by environmental interference, equipment aging, and historical bias by constructing a dynamic reliability assessment model that integrates three-dimensional factors such as road environment, equipment status, and verification history;

[0009] How to achieve automated resolution of multi-source road features based on the credibility dimension and generation of decision sets with credibility labels, and combine regional hierarchical response, equipment priority and credibility labels to execute precise resource scheduling, so as to solve the problem of decision confusion and rigid resource scheduling caused by multi-source data conflicts.

[0010] To achieve the above objectives, the present invention aims to provide a road traffic intelligent operation and maintenance system based on multi-source data fusion, comprising a credibility calculation module, a road feature conflict resolution module, and a resource credibility hierarchical scheduling module, wherein:

[0011] A three-dimensional credibility assessment model is constructed using a credibility calculation module, encompassing three dimensions: road environment factors (meteorological conditions, road surface physical conditions, traffic load characteristics), equipment status factors (operational stability, fault records, maintenance history), and verification factors (historical maintenance task success rate, timeliness deviation, problem recurrence rate). These three types of factors are mapped to the environmental, equipment, and verification dimensions, respectively. A multidimensional integration model using Bayesian networks is employed to analyze the complex nonlinear interaction relationships between these factors in real time. After standardization preprocessing to eliminate dimensional differences, a dynamic weighting strategy is used to calculate the synergistic influence coefficients of the three dimensions. Finally, the output values ​​of the three dimensions are fused using a weighted linear integration method to generate a continuously quantified dynamic credibility index that is dynamically updated in real time according to the status of various factors. This overcomes the limitations of a single indicator and provides a comprehensive, real-time, and quantitative data reliability indicator.

[0012] The credibility calculation module automatically reallocates the weights of road environment factors, equipment status factors, and verification factors in the credibility three-dimensional evaluation model through iterative learning based on the real maintenance execution feedback data provided by the verification factors (such as the deviation between task execution results and model predictions). This enables the evaluation model to have self-learning and evolution capabilities, continuously improving prediction accuracy.

[0013] The road feature conflict resolution module, through its decision matrix construction unit, automatically detects feature conflicts in multi-source data, such as road geometric conflicts, topological connection logic conflicts, and attribute hierarchy contradictions, using spatial feature analysis algorithms. Its core innovation lies in accurately mapping the dynamic credibility index generated by the credibility calculation module to the corresponding credibility dimension based on the type of conflict feature (e.g., environment-related conflicts are associated with environmental factor dimension credibility values, geometric conflicts with equipment status factor dimension credibility values, and historical data conflicts with verification factor dimension credibility values). Through this mapping mechanism, each identified conflict feature is assigned a credibility value inherited from a specific dimension, thereby generating a road conflict decision matrix with credibility labels. This translates the abstract credibility index to specific conflict points, quantifying the severity or reliability of the conflict.

[0014] The fusion unit in the road feature conflict resolution module employs a weighted evidence fusion algorithm. It aggregates meteorological-related conflict evidence in the environmental credibility dimension, integrates sensor source data conflict evidence in the equipment credibility dimension, and fuses historical maintenance feedback conflict evidence in the verification credibility dimension. In this way, conflict evidence from different credibility dimensions is fused to generate a road anomaly decision set containing three types of reliability labels: "environmental credibility label," "equipment credibility label," and "verification credibility label." This comprehensively resolves conflicts and provides clear, hierarchical, and reliability-labeled decision-making basis for subsequent scheduling.

[0015] The resource reliability grading and scheduling module, based on a dynamic reliability index, uses a continuous function segmentation mechanism (with a set threshold) to dynamically divide the entire road area into three levels: high-reliability response zone, medium-reliability response zone, and low-reliability response zone. This allows for a macroscopic identification of the overall reliability level of the area and guides the resource deployment framework.

[0016] The generation of scheduling strategies strictly follows a triple constraint, which includes:

[0017] The total amount and type of resources are determined based on the divided response zone levels (e.g., high-confidence zones are equipped with rapid response units to pursue efficiency, while low-confidence zones are deployed with enhanced inspection teams to strengthen investigations); the coverage and inspection order of key equipment are sorted by referring to the equipment weight priority parameters stored in the road conflict decision matrix; and fine filtering and processing are carried out by combining the credibility labels (environment / equipment / verification) of various anomalies in the road anomaly decision set (e.g., anomalies with verification credibility labels below a certain threshold are downgraded or given priority for review); thereby achieving precise resource allocation according to three levels: overall regional reliability level, equipment criticality, and individual credibility of specific anomalies.

[0018] The resource trust-based hierarchical scheduling module constructs a resource-decision set association graph. Its rule is to use each abnormal entry in the road anomaly decision set as a node, and the maintenance resource type as the edge attribute connecting these nodes. The graph automatically establishes dynamic associations, where abnormal nodes in the high-trust response zone are preferentially connected to automated maintenance resource units; abnormal nodes carrying high equipment trust labels are automatically associated with professional testing equipment resources; and nodes with verification trust warning labels are bound to a dedicated review and verification team, thereby intuitively displaying the matching relationship between resources and specific decision items, improving scheduling transparency and efficiency.

[0019] The second objective of this invention is to provide a method for a road traffic intelligent operation and maintenance system that integrates multi-source data, comprising the following steps:

[0020] S1. Construct a three-dimensional reliability assessment model for road operation and maintenance. The input parameters of the three-dimensional reliability assessment model include road environment factors, equipment status factors and verification factors. Using the three-dimensional reliability assessment model, a dynamic reliability index is generated through multi-factor coupling. The weights of the three-dimensional reliability assessment model are corrected in real time based on the maintenance execution feedback of the verification factors.

[0021] S2. Identify road topology conflict features in multi-source data, map them to the feature dimensions corresponding to the road topology conflict features through dynamic credibility index, construct a road conflict decision matrix, and fuse them according to feature dimensions to generate a road anomaly decision set with credibility labels.

[0022] S3. Based on the dynamic credibility index, the road area is divided into three-level response zones. Combining the equipment weight priority in the road conflict decision matrix and the credibility label in the road anomaly decision set, credibility hierarchical scheduling is performed, and a resource-decision set association graph is generated.

[0023] Compared with the prior art, the beneficial effects of the present invention are:

[0024] This invention aims to improve the intelligence level and efficiency of intelligent operation and maintenance of road traffic. First, by generating a three-dimensional credibility model and dynamic index, it achieves the quantification and dynamic evaluation of the reliability of multi-source data, providing a solid and reliable foundation for subsequent decision-making. Second, feature conflict resolution and the generation of decision sets with credibility labels efficiently solve the problem of multi-source data conflict, improving the objectivity and accuracy of decision-making basis. Finally, based on credibility classification and triple constraints, resource scheduling and correlation graphs enable precise, dynamic, and on-demand allocation of maintenance resources, significantly optimizing resource utilization, improving the timeliness and overall efficiency of operation and maintenance response, and effectively ensuring road safety and smooth traffic. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the overall module unit of the present invention;

[0026] Figure 2 This is a schematic diagram of the overall method steps of the present invention.

[0027] In the diagram: 100, Credibility Calculation Module; 200, Road Feature Conflict Resolution Module; 201, Decision Matrix Construction Unit; 202, Fusion Unit; 300, Resource Credibility Hierarchical Scheduling Module. Detailed Implementation

[0028] To better understand the purpose, technical solution, and advantages of this application, the application is described and illustrated below in conjunction with the accompanying drawings and embodiments.

[0029] Unless otherwise defined, the technical or scientific terms used in this application shall have the general meaning understood by one of ordinary skill in the art to which this application pertains. Words such as “a,” “an,” “an,” “the,” “the,” and “these” used in this application do not indicate quantitative limitation and may be singular or plural. The terms “comprising,” “including,” “having,” and any variations thereof used in this application are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or modules (units) is not limited to the listed steps or modules (units) but may include steps or modules (units) not listed, or may include other steps or modules (units) inherent to these processes, methods, products, or devices. Words such as “connected,” “linked,” and “coupled” used in this application are not limited to physical or mechanical connections but may include electrical connections, whether direct or indirect. “Multiple” used in this application refers to two or more. “And / or” describes the relationship between related objects, indicating that three relationships may exist; for example, “A and / or B” can represent: A alone, A and B simultaneously, and B alone. Normally, the character " / " indicates that the objects before and after it are in an "or" relationship. The terms "first," "second," "third," etc., used in this application are merely to distinguish similar objects and do not represent a specific order of objects.

[0030] Next, please refer to Figure 1 One of the objectives of this embodiment is to provide a road traffic intelligent operation and maintenance system that integrates multi-source data, including a credibility calculation module 100, a road feature conflict resolution module 200, and a resource credibility hierarchical scheduling module 300.

[0031] The core task of the credibility calculation module 100 is to construct a three-dimensional credibility assessment model for road operation and maintenance. This model is constructed through the following steps:

[0032] Input parameter mapping: Mapping road environment factors to the environment dimension, equipment status factors to the equipment dimension, and verification factors to the verification dimension;

[0033] Nonlinear Relationship Analysis: Real-time analysis of nonlinear interaction relationships between factors using a multidimensional fusion model of Bayesian networks;

[0034] Standardization and weighting: Standardize the factors of each dimension to eliminate differences in units, and calculate the synergistic impact coefficient of the environment / equipment / verification dimension based on a dynamic weighting strategy;

[0035] Generation of dynamic credibility index: The three-dimensional output values ​​are fused using a weighted linear integration method to generate the dynamic credibility index.

[0036] The three-dimensional reliability assessment model quantifies the reliability performance of a road system during operation and maintenance by integrating multi-dimensional input parameters. Specifically, the input parameters include road environment factors, equipment status factors, and verification factors, among which:

[0037] Road environmental factors include meteorological conditions (including the effects of temperature and precipitation), road surface physical conditions (such as the degree of cracks or potholes), and traffic load characteristics. These factors are collected in real time through a sensor network and preprocessed into structured data.

[0038] Equipment status factors focus on the operational health of road monitoring equipment (such as cameras or sensors), including its operational stability, fault records, and maintenance history, to ensure the accuracy of its output data;

[0039] Validation factors are actual feedback data from historical maintenance execution records, such as the success rate of past maintenance tasks, timeliness deviation, and problem recurrence rate, which serve as the benchmark for calibrating the three-dimensional credibility assessment model.

[0040] In the construction phase of the three-dimensional credibility assessment model, road environment factors, equipment status factors, and verification factors are mapped to three preset dimensions (environment, equipment, and verification) to form a coupled framework. This framework uses a weighted linear integration method, combined with interaction analysis between factors, to generate a dynamic credibility index. The dynamic credibility index, through a multi-factor coupling algorithm, fuses factor values ​​in real time into continuously changing quantitative indicators to reflect the dynamic fluctuation state of the overall road operation and maintenance credibility. The specific implementation process of generating the dynamic credibility index through multi-factor coupling is as follows:

[0041] Road environment factors, equipment status factors, and verification factors are input into the three-dimensional credibility evaluation model. Through a pre-defined factor-dimensional mapping mechanism, a coupling algorithm based on a multi-dimensional integration model, namely a Bayesian network, is used to analyze the nonlinear interaction relationships between factors in real time. Specifically, this includes:

[0042] First, through a pre-defined mapping mechanism, road environmental factors (such as weather conditions and road surface physical conditions) are categorized into the environmental dimension, equipment status factors (such as operational stability and fault records) into the equipment dimension, and verification factors (such as historical maintenance success rates) into the verification dimension, forming a three-dimensional structured framework. Next, a multi-dimensional integration model constructed using Bayesian networks is used to analyze the complex nonlinear interactions between these factors in real time. For example, severe weather in the environmental factors may affect the sensor accuracy of the equipment factors, while the historical problem recurrence rate in the verification factors is correlated with the changing trends of the environmental factors. This model dynamically captures the coupling effects between factors (such as the chain reaction of sudden weather changes leading to an increase in equipment failure rates) through a probabilistic inference mechanism, ensuring that these interactive effects can be identified and quantified in real time during road operation and maintenance, thereby providing basic data support for credibility assessment.

[0043] During the coupling process, the factors of each dimension are first standardized to eliminate dimensional differences. Then, the synergistic influence coefficient is calculated based on a dynamic weighting strategy, and the three-dimensional output values ​​are fused using a weighted linear integration method. Specifically, this includes:

[0044] After standardization preprocessing eliminates the dimensional differences among the factors, a dynamic weighting strategy is used to calculate the synergistic influence coefficient among the environmental, equipment, and verification dimensions. Specifically, the strategy dynamically adjusts the weight ratios based on the real-time fluctuation characteristics of the dimensional factors (such as the frequency of weather changes in the environmental dimension, the failure trend in the equipment dimension, and the magnitude of feedback deviation in the verification dimension). For example, when the state fluctuations in the equipment dimension are significant, the system automatically increases its weight ratio, and vice versa, to reflect the contribution of each dimension to the overall credibility. This weighting process is continuously optimized through learnable parameters (iteratively adjusted based on maintenance feedback data) to ultimately generate a comprehensive synergistic coefficient. This coefficient integrates the output values ​​of the environmental, equipment, and verification dimensions, ensuring accurate quantification of the reliability impact of multidimensional data in road anomaly decision-making.

[0045] The final result is a dynamic credibility index that reflects the real-time reliability of the road operation and maintenance system. This index is output in a continuously quantified and fluctuating form and is updated every minute according to the changes in the state of input factors (such as sudden weather changes or equipment failure alarms), forming a dynamic credibility indicator that can drive subsequent decision-making.

[0046] The specific process of multi-factor coupling involves inputting road environment factors, equipment status factors, and verification factors into the three-dimensional credibility assessment model. Through a pre-defined factor-dimensional mapping mechanism, a coupling algorithm based on a multi-dimensional integration model (i.e., Bayesian network) is used to analyze the nonlinear interaction relationships between factors in real time. During the coupling process, the factors of each dimension are first standardized and preprocessed to eliminate dimensional differences. Then, the dimensional synergistic influence coefficient is calculated based on a dynamic weighting strategy, and the three-dimensional output values ​​are fused using a weighted linear integration method. Finally, a dynamic credibility index is generated.

[0047] To enhance the adaptability of the 3D credibility assessment model, the credibility calculation module 100 incorporates a real-time correction mechanism. This mechanism dynamically adjusts the weight parameters of the 3D credibility assessment model based on maintenance execution feedback provided by validation factors. Specifically, when maintenance execution feedback data (such as the latest task report or abnormal event feedback) is received, the credibility calculation module 100 triggers a weight correction process by comparing the deviation between the feedback results and the actual model predictions. This process uses iterative learning methods (such as backpropagation or online reinforcement learning techniques) to re-evaluate the weight allocation of each factor. For example, if maintenance feedback shows that the impact of environmental factors on a specific road segment is underestimated, the manually adjustable weight adjustment coefficient is automatically increased, making the 3D credibility assessment model more sensitive to environmental changes. Conversely, if validation factor feedback proves that the contribution of equipment status factors is too high, the 3D credibility assessment model reduces the weight proportion of that dimension. This closed-loop feedback process ensures the adaptability and real-time optimization capability of the 3D credibility assessment model, enabling it to continuously adapt to the complex and changing environment of road maintenance, ultimately enhancing the predictive accuracy and decision support effectiveness of the dynamic credibility index. The specific implementation of the weight correction process of the 3D credibility assessment model is as follows:

[0048] When the maintenance execution feedback data provided by the validation factors (including the latest task completion report, abnormal event handling records, or problem recurrence status) is input into the credibility calculation module 100, the deviation magnitude and distribution characteristics between the feedback data and the model's existing prediction results are automatically compared. This triggers a weight redistribution process based on iterative learning, where the iterative learning method uses backpropagation or online reinforcement learning techniques to analyze the weight deviation contribution of road environmental factors, equipment status factors, and validation factors through a gradient optimization mechanism. In the weight correction stage, if the deviation analysis indicates that the impact of environmental factors is underestimated (e.g., continuous meteorological disasters were not effectively warned), the weight adjustment coefficient of the environmental dimension is automatically increased. If the validation factor feedback confirms that the credibility assessment of the equipment status factor is too high (e.g., abnormal data output by faulty equipment is not identified), the weight ratio of the equipment dimension is reduced. This closed-loop process uses maintenance events as the driving unit and outputs optimized weight parameters in real time after each feedback is received, ensuring that the credibility three-dimensional evaluation model continuously adapts to the dynamic changes in the road operation and maintenance scenario.

[0049] The core function of the decision matrix construction unit 201 in the road feature conflict resolution module 200 is to solve the topological inconsistency problem in multi-source road data through a systematic approach, and to assign reliability labels to abnormal decisions based on a reliable metric system.

[0050] The decision matrix construction unit 201 first identifies road topology conflict features in multi-source data (such as remote sensing images, vehicle sensor streams, and historical road network topology databases) using a spatial feature analysis algorithm. Typical scenarios include conflicting meteorological data, conflicting road segment geometry (such as lane line position offsets), conflicting topological connection logic (such as missing intersection associations), or conflicting attribute levels (such as speed limit information conflicts). During the identification process, the dynamic credibility index is mapped to the feature dimensions corresponding to various road topology conflict features. Conflicts related to environmental conditions (such as conflicting meteorological data or road condition judgments) are associated with the credibility value of the environmental factor dimension; for geometric conflicts, the credibility value of the equipment status factor dimension is associated; for historical data conflicts, the credibility value of the verification factor dimension is associated. Through this mapping mechanism, each conflict feature automatically inherits the credibility value of a specific dimension, forming a three-dimensional road conflict decision matrix. This matrix quantifies and records the type, spatial distribution, and corresponding multi-dimensional reliability indicators of the conflict features. This three-dimensional matrix includes environmental credibility, equipment credibility, and verification credibility.

[0051] In the decision fusion stage, the fusion unit 202 in the road feature conflict resolution module 200 adopts a feature dimension-oriented weighted evidence fusion algorithm to adaptively aggregate the conflict features of the same type in the decision matrix according to their credibility weights. The specific process is as follows:

[0052] This system aggregates conflicting evidence related to meteorological factors in the environmental credibility dimension, integrates conflicting evidence from sensor source data in the equipment credibility dimension, and fuses conflicting evidence from historical maintenance feedback in the verification credibility dimension. Each aggregation unit automatically generates a triplet of "anomaly type - location coordinates - multi-dimensional credibility score," ultimately synthesizing a road anomaly decision set with credibility labels. The innovation of this decision set lies in attaching three types of reliability tags—environmental credibility label, equipment credibility label, and verification credibility label—to each anomaly conclusion. This allows downstream modules to finely assess the credibility level of anomaly decisions based on multi-dimensional reliability scores, providing a tiered response basis for subsequent maintenance resource scheduling. The specific process for generating the road anomaly decision set is as follows:

[0053] In the environmental credibility dimension, we aggregate conflicting meteorological correlation evidence, extract the correlation between meteorological data and road anomaly characteristics (such as inconsistent judgments on road slipperiness by multiple sensors during heavy rain), quantify the matching degree between meteorological parameters and anomaly reports through Pearson correlation coefficient, and use the matching deviation value as the weight of conflicting evidence in the environmental dimension.

[0054] In the dimension of device credibility, conflict evidence from sensor source data is integrated, and differences in geometric shape or state judgment of different monitoring devices (such as cameras and radar) on the same road segment are analyzed (such as lane line position deviation exceeding the threshold). Evidence credibility weights are assigned based on the historical accuracy records of the devices.

[0055] In the dimension of verifying credibility, conflicting evidence from historical maintenance feedback is integrated, and the handling results of similar cases in the current abnormal characteristics are compared with those in the historical maintenance records (such as the recurrence rate of pothole problems). The historical verification success rate is used as the confidence benchmark for conflicting evidence.

[0056] The resource trust level scheduling module 300 divides the entire road network into three levels of response areas: high trust response area, medium trust response area, and low trust response area based on the dynamic trust index. The division logic adopts a continuous function segmentation mechanism based on the dynamic trust index threshold.

[0057] The resource trustworthy hierarchical scheduling module 300 synchronously accesses the equipment weight priority parameters (i.e., the key equipment influence coefficients marked by equipment status factor dimensions in the road conflict decision matrix) in the road conflict decision matrix, as well as the trustworthy label (including environmental trustworthiness / equipment trustworthiness / verification trustworthiness score) of each anomaly item in the road anomaly decision set, and performs trustworthy hierarchical scheduling, specifically including:

[0058] First, determine the total amount of resources to be deployed according to the response zone level (equip the high-confidence zone with rapid response units and deploy enhanced inspection teams in the low-confidence zone). Second, sort the coverage order of key equipment according to the equipment weight priority (e.g., prioritize the repair of high-weight sensor failures). Finally, combine the trust label filtering to make low reliability decisions (e.g., downgrade anomalies with verification trust labels below the threshold).

[0059] During the hierarchical scheduling implementation phase, the resource trust hierarchical scheduling module 300 uses a multi-objective optimization engine to dynamically generate a resource-decision set association graph. This graph uses entries from the road anomaly decision set as nodes and maintenance resource type (manpower / equipment / vehicles) as edge attributes, and establishes a mapping through the following rules:

[0060] Anomaly nodes in the high-reliability response zone are connected to automated maintenance resource units;

[0061] Nodes carrying high-reliability equipment tags are associated with professional testing equipment resources;

[0062] The node with the verification credibility warning label is bound to the verification team for review;

[0063] The mapping process integrates the topological feature coordinates of the conflict decision matrix to form a four-element relationship network of "spatial location-anomaly type-resource type-trust dimension", and finally outputs a visualized scheduling scheme to provide maintenance departments with decision support that combines reliable traceability and dynamic resource adaptation.

[0064] As can be seen from the above description, the road traffic intelligent operation and maintenance system with multi-source data fusion provided in this embodiment has the following technical effects:

[0065] The credibility calculation module 100 integrates road environmental factors, equipment status factors, and verification factors to construct a three-dimensional evaluation model, generating a credibility index that is dynamically updated on a minute-by-minute basis. The model weights are then corrected through a maintenance feedback loop, addressing the dynamic fluctuations in reliability caused by environmental interference, equipment aging, and historical biases in multi-source data. The road feature conflict resolution module 200 identifies and quantifies road topology, geometry, and historical logical conflicts based on the credibility index dimension mapping. It constructs a three-dimensional conflict decision matrix labeled with environmental credibility, equipment credibility, and verification credibility. A weighted evidence fusion algorithm is used to generate a road anomaly decision set with triple credibility labels, overcoming multiple challenges. The resource trust-based hierarchical scheduling module 300 addresses the problem of conflicting source data features leading to chaotic decision-making criteria. Based on a trust index, it divides response areas into high, medium, and low levels. Combining the device weight priority in the conflict matrix with the trust labels of the anomaly decision set, it implements a triple-constraint scheduling strategy: differentiated resource allocation by response area level, execution order sorted by device criticality, and filtering of low-reliability anomalies based on trust labels. Ultimately, it constructs a four-element resource decision association graph integrating spatial location, anomaly type, resource type, and trust dimension, achieving precise matching and dynamic adaptation of maintenance resources. This eliminates the rigid mismatch bottleneck of traditional scheduling mechanisms, significantly improving operational efficiency and decision reliability.

[0066] Please see Figure 2 The second objective of this embodiment is to provide a method for a road traffic intelligent operation and maintenance system that integrates multi-source data, the steps of which are as follows:

[0067] S1. Construct a three-dimensional reliability assessment model for road operation and maintenance. The input parameters of the three-dimensional reliability assessment model include road environment factors, equipment status factors and verification factors. Using the three-dimensional reliability assessment model, a dynamic reliability index is generated through multi-factor coupling. The weights of the three-dimensional reliability assessment model are corrected in real time based on the maintenance execution feedback of the verification factors.

[0068] S2. Identify road topology conflict features in multi-source data, map them to the feature dimensions corresponding to the road topology conflict features through dynamic credibility index, construct a road conflict decision matrix, and fuse them according to feature dimensions to generate a road anomaly decision set with credibility labels.

[0069] S3. Based on the dynamic credibility index, the road area is divided into three-level response zones. Combining the equipment weight priority in the road conflict decision matrix and the credibility label in the road anomaly decision set, credibility hierarchical scheduling is performed, and a resource-decision set association graph is generated.

[0070] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A road traffic intelligent operation and maintenance system based on multi-source data fusion, characterized in that, The application comprises a credibility calculation module (100), a road feature conflict resolution module (200) and a resource credibility hierarchical scheduling module (300), wherein: The credibility calculation module (100) constructs a credibility three-dimensional evaluation model for road operation and maintenance, wherein the input parameters of the credibility three-dimensional evaluation model comprise road environment factors, device state factors and verification factors, the dynamic credibility index is generated by multi-factor coupling through the credibility three-dimensional evaluation model, and the weight of the credibility three-dimensional evaluation model is corrected in real time based on the maintenance execution feedback of the verification factors; The road feature conflict resolution module (200) identifies road topology conflict features in multi-source data, maps the dynamic credibility index to the feature dimension corresponding to the road topology conflict features, constructs a road conflict decision matrix, and generates a road anomaly decision set with a credibility label through fusion according to the feature dimension; the road feature conflict resolution module (200) comprises a decision matrix construction unit (201), which is used to generate a road conflict decision matrix, and specifically comprises: The spatial feature analysis algorithm is used to detect road geometric morphology conflicts, topology connection logic conflicts and attribute level contradictions; the dynamic credibility index is mapped to the feature dimension corresponding to the conflict features, wherein the credibility value of the environment state related conflict is associated with the environment factor dimension, the credibility value of the geometric morphology conflict is associated with the device state factor dimension, and the credibility value of the historical data conflict is associated with the verification factor dimension; through the above mechanism of mapping the dynamic credibility index to the feature dimension corresponding to the conflict features, each conflict feature automatically inherits the credibility value of the specific dimension, and a road conflict decision matrix is generated; The resource credibility hierarchical scheduling module (300) divides the road area into three response zones based on the dynamic credibility index, combines the device weight priority in the road conflict decision matrix and the credibility label in the road anomaly decision set, performs credibility hierarchical scheduling, and generates a resource-decision set association graph.

2. The multi-source data fusion road traffic intelligent operation and maintenance system according to claim 1, characterized in that, The road environment factors comprise meteorological conditions, road surface physical conditions and traffic load characteristics; The device state factors comprise the working stability of the road monitoring device, fault records and maintenance history; the verification factors comprise the success rate of historical maintenance tasks, time deviation and problem recurrence rate. 3.The multi-source data fusion road traffic intelligent operation and maintenance system according to claim 1, characterized in that, The generation process of the dynamic credibility index specifically comprises: The road environment factors are mapped to the environment dimension, the device state factors are mapped to the device dimension, and the verification factors are mapped to the verification dimension; the non-linear interaction relationship between the factors is analyzed in real time through a multi-dimensional integration model of Bayesian network; after standardization preprocessing to eliminate dimension differences, the synergistic influence coefficients of the environment dimension, the device dimension and the verification dimension are calculated based on a dynamic weighting strategy; the three-dimensional output values are fused by using a weighted linear integration method to generate a dynamic credibility index which is continuously quantized and dynamically updated according to the factor state. 4.The multi-source data fusion road traffic intelligent operation and maintenance system according to claim 1, characterized in that, The weight correction process of the credibility three-dimensional evaluation model specifically comprises: According to the deviation between the maintenance execution feedback data provided by the verification factors and the actual model prediction, the weights of the road environment factors, the device state factors and the verification factors are redistributed through an iterative learning method. 5.The multi-source data fusion road traffic intelligent operation and maintenance system according to claim 1, characterized in that, The road feature conflict resolution module (200) comprises a fusion unit (202) configured to generate a road anomaly decision set, and specifically comprises: A weighted evidence fusion algorithm is adopted to aggregate meteorological correlation conflict evidence in the environmental credibility dimension, integrate sensor source data conflict evidence in the equipment credibility dimension, and fuse historical maintenance feedback conflict evidence in the verification credibility dimension, so as to generate a road anomaly decision set containing three types of reliability labels, i.e., an environmental credibility label, an equipment credibility label and a verification credibility label. 6.The multi-source data fusion road traffic intelligent operation and maintenance system according to claim 1, characterized in that, The three-level response area is divided by using a continuous function segmentation mechanism of a dynamic credibility index threshold, and the three-level response area comprises a high credibility response area, a medium credibility response area and a low credibility response area. 7.The multi-source data fusion road traffic intelligent operation and maintenance system according to claim 1, characterized in that, The construction rule of the resource-decision set association graph is: The entries of the road anomaly decision set are taken as nodes, and the maintenance resource types are taken as edge attributes; The abnormal nodes in the high credibility response area are connected to the automatic maintenance resource units, the nodes carrying the high equipment credibility label are associated with the professional detection equipment resources, and the nodes with the verification credibility warning label are bound to the review verification team. 8.The multi-source data fusion road traffic intelligent operation and maintenance system according to claim 1, characterized in that, The process of the resource credibility hierarchical scheduling module (300) performing the credibility hierarchical scheduling comprises: The total amount of resources is determined according to the response area level, the high credibility response area is equipped with a rapid response unit, and the low credibility response area is deployed with an intensive inspection group; The key equipment coverage order is sorted according to the equipment weight priority parameter; and the low reliability abnormal items are filtered according to the credibility label, and the abnormal items with a verification credibility label lower than a threshold are subjected to degradation processing. 9.A method using a road traffic intelligent operation and maintenance system comprising the multi-source data fusion of any one of claims 1-8, characterized in that, The method comprises the following steps: S1, a credibility three-dimensional evaluation model for road operation and maintenance is constructed, wherein the input parameters of the credibility three-dimensional evaluation model comprise road environment factors, equipment state factors and verification factors, a dynamic credibility index is generated by using the credibility three-dimensional evaluation model through multi-factor coupling, and the weights of the credibility three-dimensional evaluation model are corrected in real time based on the maintenance execution feedback of the verification factors; S2, road topological conflict features in multi-source data are identified, the dynamic credibility index is mapped to the feature dimension corresponding to the road topological conflict features, a road conflict decision matrix is constructed, and a road anomaly decision set with a credibility label is generated by fusion according to the feature dimension; The process of constructing the road conflict decision matrix comprises: A spatial feature analysis algorithm is used to detect road geometric shape conflicts, topological connection logic conflicts and attribute level contradictions; The dynamic credibility index is mapped to the feature dimension corresponding to the conflict features, wherein the credibility value of the environment state related conflict is associated with the environment factor dimension, the credibility value of the geometric shape conflict is associated with the equipment state factor dimension, and the credibility value of the historical data conflict is associated with the verification factor dimension; through the above mechanism of mapping the dynamic credibility index to the feature dimension corresponding to the conflict features, each conflict feature automatically inherits the credibility value of the specific dimension, and a road conflict decision matrix is generated; S3, based on the dynamic credibility index, a road area is divided into three response areas, the equipment weight priority in the road conflict decision matrix and the credibility label in the road anomaly decision set are combined, credibility hierarchical scheduling is performed, and a resource-decision set association graph is generated.

Images