A method and system for quantitatively evaluating and early warning of risks of multi-modal transport path links

Abstract

The application belongs to the technical field of multimodal transport, and specifically discloses a kind of quantitative evaluation and early warning method and system of multimodal transport path link coupling risk, the application quantifies the influence of time coupling, space coupling and resource coupling on risk through the coupling relationship modeling method between multimodal transport links, reveals the dynamic coupling risk generated by the interaction between each link in the multimodal transport path, improves the comprehensiveness and accuracy of risk identification;By real-time, dynamic and quantitative evaluation of the coupling risk between each link in the multimodal transport path, a multi-dimensional, quantitative risk assessment result is provided, the resilience and efficiency of multimodal transport operation are improved, and the potential economic loss and operation interruption risk is reduced;By continuously receiving real-time data and dynamically updating the risk assessment result, the risk early warning capability of the multimodal transport path is improved, which is beneficial to the quick response and response measures of the management personnel, and reduces the probability and influence of risk events.

Description

Technical Field

[0001] This invention belongs to the field of multimodal transport technology, specifically relating to a method and system for quantitative assessment and early warning of coupling risks in multimodal transport routes. Background Technology

[0002] Multimodal transport, as an important form of transportation organization, organically combines various modes of transport such as sea, land, and air, achieving seamless connection of goods from origin to destination. It holds irreplaceable strategic significance for reducing logistics costs, improving transportation efficiency, and reducing environmental pollution. However, the complexity of multimodal transport lies in its involvement of multiple independent operating entities, various modes of transport, multiple transfer nodes, and complex collaborative operation processes. Various internal and external risk factors may exist at each stage. These factors are not isolated but interconnected and mutually influential, forming complex coupling effects. Timely and accurate identification and assessment of these coupled risks are crucial to ensuring the safe and efficient operation of the multimodal transport system and enhancing its resilience and adaptability. The lack of an effective risk identification mechanism will directly threaten the stability and reliability of the entire supply chain.

[0003] Existing technologies have many shortcomings in multimodal transport risk identification: On the one hand, existing methods for identifying multimodal transport risks often focus on risk assessment of a single transport link, making it difficult to effectively capture the complex interactions and coupling effects between different modes of transport and different operating entities, resulting in the locality and one-sidedness of risk analysis.

[0004] On the other hand, most existing methods for identifying risks in multimodal transport rely on expert experience or historical data. When faced with dynamically changing transport environments and unforeseen events, these methods are slow to respond, poorly adaptable, and struggle to provide early warnings and quantitative assessments of potential coupled risks. Information barriers further complicate data sharing and integration across different stages, exacerbating the difficulty of risk identification. Summary of the Invention

[0005] In view of this, in order to solve the problems mentioned in the background technology, a quantitative assessment and early warning method and system for the coupling risks of multimodal transport routes is proposed.

[0006] The objective of this invention can be achieved through the following technical solution: The first aspect of this invention provides a method for quantitative assessment and early warning of coupling risks in multimodal transport routes, including: S1, data acquisition: collecting and preprocessing link data and environmental perception data of multimodal transport routes through Internet of Things sensing devices.

[0007] S2. Construction of digital twin: Map process data and environmental perception data to the corresponding nodes in the constructed multimodal transport path environment network model to construct a multimodal transport digital twin.

[0008] S3. Dynamic Coupling Coefficient Calculation: Based on the multimodal transport digital twin, define and calculate the real-time coupling coefficient between each pair of adjacent links in the multimodal transport path.

[0009] S4. Coupling Risk Identification and Assessment: Based on the real-time coupling coefficient, a dynamic risk map of the multimodal transport route is constructed in the multimodal transport digital twin, and the real-time coupling coefficients of several adjacent links of the multimodal transport route are aggregated to calculate the risk indicator system.

[0010] S5. Risk Visualization: Based on the multimodal transport digital twin, real-time coupling coefficient and risk indicator system, a multimodal transport route environment coupling risk identification report is generated. When the risk indicators exceed the preset threshold, multi-level early warning information is generated and visualized.

[0011] A second aspect of the present invention provides a multimodal transport route link coupling risk identification system, comprising: a data acquisition module, a digital twin construction module, a dynamic coupling coefficient calculation module, a coupling risk identification and assessment module, and a risk visualization display module.

[0012] The data acquisition module is connected to the digital twin construction module, the digital twin construction module is connected to the dynamic coupling coefficient calculation module, the dynamic coupling coefficient calculation module is connected to the coupling risk identification and assessment module, and the coupling risk identification and assessment module is connected to the risk visualization display module.

[0013] The data acquisition module collects and preprocesses process data and environmental perception data of multimodal transport routes through IoT sensing devices.

[0014] The digital twin construction module maps process data and environmental perception data to corresponding nodes in the constructed multimodal transport path environment network model, thereby constructing a multimodal transport digital twin.

[0015] The dynamic coupling coefficient calculation module, based on the multimodal transport digital twin, defines and calculates the real-time coupling coefficient between each pair of adjacent links in the multimodal transport path.

[0016] The coupling risk identification and assessment module constructs a dynamic risk map of multimodal transport routes in the multimodal transport digital twin based on real-time coupling coefficients, and aggregates the real-time coupling coefficients of several adjacent links of the multimodal transport route to calculate the risk indicator system.

[0017] The risk visualization module generates a multimodal transport route environment coupling risk identification report based on the multimodal transport digital twin, real-time coupling coefficient, and risk indicator system. When the risk indicators exceed the preset threshold, it generates multi-level early warning information and displays it visually.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. The present invention, through the proposed modeling method for coupling relationships between multimodal transport links, abstracts the multimodal transport path into a directed graph structure, quantifies the impact of time coupling, spatial coupling and resource coupling on risk, reveals the dynamic coupling risk generated by the interaction between various links in the multimodal transport path, and improves the comprehensiveness and accuracy of risk identification.

[0019] 2. This invention provides multi-dimensional and quantitative risk assessment results by evaluating the coupling risks between various links in the multimodal transport route in real time, dynamically and quantitatively, making risk management more refined and scientific, improving the resilience and efficiency of multimodal transport operations, and reducing potential economic losses and operational interruption risks.

[0020] 3. This invention continuously receives real-time data and dynamically updates risk assessment results. When risk indicators exceed preset thresholds, it promptly triggers early warnings, thereby enhancing the risk warning capabilities of multimodal transport routes. This helps managers respond quickly and take countermeasures, effectively reducing the probability and impact of risk events. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram illustrating the implementation steps of the method of the present invention.

[0023] Figure 2 This is a schematic diagram of the system module connections of the present invention. Detailed Implementation

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] Example 1

[0026] Please see Figure 1As shown, the present invention provides a method for quantitative assessment and early warning of coupling risks in multimodal transport routes. The specific steps are as follows: S1, Data acquisition: Collect and preprocess link data and environmental perception data of multimodal transport routes through IoT sensing devices.

[0027] In one feasible embodiment of the present invention, the multimodal transport route data includes basic information about the transport links, transport resource status data, time dimension data, and cost and efficiency data.

[0028] In one specific example, the basic information of the transportation links is used to describe the inherent attributes and topological relationships of each transportation link to which the multimodal transport route belongs, including but not limited to the type and sub-type of transportation mode, hub node information, route topological relationship, and cargo attributes.

[0029] For example, the transportation mode types include, but are not limited to, road, rail, sea, air, and pipeline; the sub-types include, but are not limited to, ordinary trucks and refrigerated trucks on roads, and container ships and bulk carriers on sea; the hub node information includes, but is not limited to, port codes, station numbers, warehouse IDs, geographical coordinates (latitude and longitude), and node types, such as transit hubs and destinations; the path topology includes, but is not limited to, origin and destination codes, connection methods between adjacent links, transportation route numbers, and road, waterway, and railway line grades; the cargo attributes include, but are not limited to, cargo type codes, packaging types, weight, volume, hazard class, and temperature control requirements.

[0030] The transportation resource status data reflects the real-time operational status of transportation vehicles and supporting resources, including but not limited to the status of transportation vehicles, hub equipment, and human resources.

[0031] For example, the status of the means of transport includes, but is not limited to, the ID, real-time location, speed, load factor, engine speed, fuel consumption, fault codes, tire pressure (for highways), draft (for sea transport), and track gauge matching degree (for railways) of vehicles, ships, and trains; the status of hub equipment includes, but is not limited to, the utilization rate of port cranes, the occupancy rate of container storage spaces, the status of railway turnouts, the operating parameters of warehouse temperature control equipment, and the failure rate of security inspection equipment; the status of human resources includes, but is not limited to, the on-duty status, qualification level, continuous working hours, and fatigue score of drivers, crew members, and dispatchers.

[0032] The time dimension data is used to record the time nodes and time consumption of each stage, including but not limited to planned time, actual time, and time constraints.

[0033] For example, the planned time includes, but is not limited to, departure, takeoff, arrival at port and station, cargo loading and unloading time, and transit connection interval; the actual time includes, but is not limited to, actual departure, takeoff, arrival at port and station, actual loading and unloading time, delay time and reason code; the time constraints include, but are not limited to, the time requirements stipulated in the transportation contract, port and station operation time windows, and traffic control time periods.

[0034] The cost and efficiency data are used to quantify the economic attributes and operational efficiency of the transportation process, including but not limited to cost data and efficiency data.

[0035] For example, the cost data includes, but is not limited to, unit mileage freight, fuel and electricity consumption costs, port and station operation fees, insurance premiums, fines such as overtime fees, and maintenance costs; the efficiency data includes, but is not limited to, transportation vehicle turnover rate, port and station loading and unloading efficiency, cargo damage rate, punctuality rate, and resource idle time.

[0036] The environmental perception data includes meteorological information, traffic conditions, policies and regulations, and social events.

[0037] One specific example is that the meteorological information includes, but is not limited to, real-time weather, weather warnings, and medium- and long-term weather forecasts.

[0038] For example, the real-time weather includes, but is not limited to, temperature, humidity, precipitation, wind force level, visibility, and weather phenomena such as sunny, rainy, snowy, and foggy; the weather warnings include, but are not limited to, rainstorm, blizzard, typhoon, and fog warning levels and impact ranges, high temperature and low temperature warning thresholds, and lightning warning probability; the medium- and long-term weather forecasts include, but are not limited to, the probability of precipitation in the next 72 hours, wind force trends, and temperature fluctuation range.

[0039] The traffic conditions mentioned include, but are not limited to, road conditions, waterway and shipping route conditions, and traffic around ports and stations.

[0040] For example, the road conditions include, but are not limited to, highway congestion index, road surface smoothness, location and impact range of construction sections, and traffic accident information, such as location, type, and processing time; the waterway and route conditions include, but are not limited to, waterway depth, water flow speed, reef warnings, sea wave level, air traffic control information, and flight delay rate; the port and station surrounding traffic includes, but is not limited to, port access road congestion, parking lot saturation around the station, and queue length at hub entrances.

[0041] The policies and regulations mentioned include, but are not limited to, traffic control policies, cross-border regulatory policies, and updates to industry standards.

[0042] For example, the traffic control policies include, but are not limited to, restricted hours, areas, load limits, and prohibited routes for the transport of dangerous goods; the cross-border regulatory policies include, but are not limited to, changes in customs clearance procedures, quarantine requirements, and tariff adjustments; and the industry standard updates include, but are not limited to, revisions to transport safety regulations, upgrades to emission standards, and changes in equipment access conditions.

[0043] The social events mentioned include, but are not limited to, public safety incidents, infrastructure incidents, and economic fluctuations.

[0044] For example, the public safety incidents include, but are not limited to, areas under epidemic control and the scope of impact of mass incidents; the infrastructure incidents include, but are not limited to, bridge and tunnel maintenance and closures, port strikes, and railway signal system failures; and the economic fluctuations include, but are not limited to, increases in fuel prices, strikes in logistics parks, and supply chain disruption warnings.

[0045] The preprocessing includes timestamp synchronization, data alignment, missing value imputation, outlier filtering, and data standardization.

[0046] It should be noted that the timestamp synchronization aims to unify the time base of different data sources, ensuring consistency in the time dimension. The data alignment aims to achieve spatial and logical matching and association of data from different dimensions. Missing value imputation aims to handle missing data caused by data acquisition interruptions or malfunctions, avoiding impact on subsequent analysis. Outlier filtering aims to identify and handle abnormal data caused by equipment failures or human error. Data standardization aims to eliminate differences in the dimensions of different data, bringing the data to a uniform scale for easier subsequent model calculations.

[0047] The timestamp synchronization is achieved through network time protocol calibration, the data alignment is achieved through time series interpolation algorithm, the missing value imputation is achieved through regression prediction model or nearest neighbor interpolation method, the outlier filtering is achieved through statistical threshold or isolated forest algorithm, and the data standardization is achieved through maximum and minimum value normalization or Z-score standardization method.

[0048] S2. Construction of digital twin: Map process data and environmental perception data to the corresponding nodes in the constructed multimodal transport path environment network model to construct a multimodal transport digital twin.

[0049] In one feasible embodiment of the present invention, the specific construction process of the multimodal transport route environment network model includes: defining each independent transport link in the multimodal transport route as a node, and assigning the node a link type, geographical coordinates, planned duration, carrier name, historical average risk level, maximum throughput capacity, and current state feature vector.

[0050] The connection between two adjacent or logically related links is defined as an edge, and the edge is assigned weights such as logistics cost, time cost, initial value of risk propagation probability, and actual distance.

[0051] Based on the defined nodes and edges, a directed graph model is constructed, and the directed graph model is stored in the form of an adjacency matrix or an adjacency list, thereby building a network model for multimodal transport path environment.

[0052] It should be noted that the technical implementation of the directed graph model is as follows: a visual directed graph is built using a graph database such as Neo4j or a network modeling tool such as Python NetworkX.

[0053] Furthermore, in the directed graph model, circular nodes represent transportation links and square nodes represent operation links; line thickness represents weight, with thicker lines indicating higher weight and stronger dependencies; and arrow direction represents process logic, such as loading / unloading node → transportation node, with the arrow pointing towards the transportation node.

[0054] In one feasible embodiment of the present invention, the specific construction process of the multimodal transport digital twin includes: placing a multimodal transport path environment network model in a digital space, and mapping the link data and environmental perception data of the multimodal transport path to the multimodal transport path environment network model in the digital space to achieve accurate mapping of the physical world, thereby constructing a multimodal transport digital twin.

[0055] A series of physical rules and business logic are injected into each node, edge, and their interactions in the digital twin.

[0056] The physical rules include kinematic rules, capacity constraint rules, equipment performance rules, and environmental impact rules, while the business logic includes process dependency logic, resource scheduling logic, timeliness constraint logic, and exception handling logic.

[0057] It should be noted that the kinematic rules are specifically defined as the motion characteristics of vehicles and ships, such as speed, acceleration, and turning radius, which, combined with real-time traffic or weather data, predict their future position and arrival time.

[0058] The capacity constraint rule is specifically defined as modeling the capacity of nodes such as port yards and warehouses and links such as roads and railway lines. For example, when the utilization rate of a port yard reaches 90%, containers arriving later will face the risk of waiting or being transferred to an alternative yard.

[0059] The equipment performance rules are specifically defined as modeling the operating efficiency, failure rate, and maintenance cycle of equipment such as cranes and loading / unloading machinery.

[0060] For example, after a crane has been working continuously for 8 hours, its failure probability will increase according to a specific curve.

[0061] The environmental impact rule is specifically defined as associating real-time weather data, such as strong winds, dense fog, and heavy rain, with physical parameters such as transportation speed, visibility, and equipment operating efficiency.

[0062] For example, when the wind speed exceeds a certain threshold, the operating efficiency of the dock crane will decrease by 20%, or operations will be suspended.

[0063] The process dependency logic is specifically defined as clarifying the sequential dependencies between various transportation links in the multimodal transport process.

[0064] For example, container unloading can only begin after the vessel has berthed and customs declaration has been completed; and land truck loading can only begin after the containers have been unloaded and transferred to the designated stacking location.

[0065] The resource scheduling logic is specifically defined as the mechanism for modeling the allocation and competition of key resources.

[0066] For example, during peak demand periods, port berths, cranes, and trucks may experience resource competition and waiting.

[0067] The timeliness constraint logic is specifically defined as the promised transportation timeliness for different types of goods or service levels.

[0068] For example, the dwell time of cold chain goods in a certain stage cannot exceed a preset threshold.

[0069] The exception handling logic specifically defines contingency plans for handling common abnormal situations.

[0070] For example, a rule that automatically switches goods to road transport when a railway train service is cancelled.

[0071] It should be further explained that the rules and logic are implemented using an event-driven programming model and managed and executed by a rule engine. When the state of any virtual entity in the digital twin changes, the corresponding rules and logic will be triggered in real time, updating the internal state of the model and providing accurate input for subsequent coupling coefficient calculations.

[0072] This invention proposes a modeling method for the coupling relationships between multimodal transport links, which abstracts the multimodal transport path into a directed graph structure, quantifies the impact of time coupling, spatial coupling and resource coupling on risk, reveals the dynamic coupling risks generated by the interaction between various links in the multimodal transport path, and improves the comprehensiveness and accuracy of risk identification.

[0073] S3. Dynamic Coupling Coefficient Calculation: Based on the multimodal transport digital twin, define and calculate the real-time coupling coefficient between each pair of adjacent links in the multimodal transport path.

[0074] In one feasible embodiment of the present invention, each pair of adjacent links in the multimodal transport path is defined as follows: The expression for the real-time coupling coefficient is defined as follows: .

[0075] The real-time coupling coefficient It is a dynamic, dimensionless value between [0,1] used to quantify upstream processes. Potential risks or abnormal conditions can be transmitted to downstream processes. and downstream links The likelihood and intensity of negative impacts on normal operation.

[0076] The real-time coupling coefficient The specific calculation formula is as follows: ,in Indicates upstream link The real-time risk status assessment value is a comprehensive indicator that reflects the process. The severity of various potential risks currently existing, such as delays, malfunctions, congestion, the impact of abnormal weather, or over-utilization of resources.

[0077] It should be noted that the above The value range is [0,1], where 0 represents no risk and 1 represents severe risk.

[0078] Indicates downstream links The real-time capacity and buffering capacity assessment value is a comprehensive indicator that reflects the process. Absorption or buffering from upstream links The ability to handle risk.

[0079] It should be noted that the above The value range is [0,1], where 0 indicates no buffering capacity or saturated capacity, and 1 indicates maximum buffering capacity or sufficient capacity.

[0080] Representation Steps With links The inherent coupling strength between them is a preset, relatively static weight value, reflecting the links based on historical data analysis and domain expert experience judgment. For the process The degree of structural dependence and the weight of influence.

[0081] It should be noted that the aforementioned steps With links Inherent coupling strength between The calculation takes into account the following inherent properties: time-coupled weights , indicating a stage Delays on the process The degree of impact on the timeliness of the plan; information coupling weight , indicating a stage Information updates on the process The degree of influence of the decision-making process; physical coupling weight , indicating a stage The physical state is a link Necessary preconditions for starting the task; resource sharing coupling weights , indicating a stage and links Do they share critical resources? The value is obtained by retrospective analysis of historical risk events, combined with training of historical data using machine learning models, and supplemented by weight adjustment and calibration by domain experts.

[0082] It is a non-linear mapping function used for synthesis , and Calculate the .

[0083] It should be noted that the above The function can be implemented in the following two ways: (1) Inference function based on fuzzy logic: by defining fuzzy sets and fuzzy rule bases.

[0084] For example, the fuzzy set specifically includes: Defined as low risk, medium risk, and high risk; Defined as high buffer, medium buffer, and low buffer.

[0085] For example, if It is high risk and If it is a low buffer, then It is highly coupled and employs a fuzzy inference mechanism, as exemplified by Mamdanior Sugeno. .

[0086] An example of the fuzzy rule base is shown in Table 1 below.

[0087] Table 1 Examples of Fuzzy Rule Bases

[0088]

[0089] (2) Machine learning-based prediction models: , and The features are input into a pre-trained machine learning model, such as a gradient boosting decision tree, neural network, or support vector machine, to directly predict... The model uses supervised learning with real-world data from historical risk events to optimize its predictive accuracy.

[0090] In one feasible embodiment of the present invention, the upstream link Real-time risk status assessment value The calculation formula is ,in Indicates upstream link Time deviation rate Upstream links Real-time fault status or fault probability of critical equipment. Indicates upstream link Real-time utilization rate of critical resources Indicates upstream link Environmental risk factors These are the set weights for time deviation rate, failure probability, real-time utilization rate, and environmental risk factor, respectively. .

[0091] It should be noted that the time deviation rate is calculated and normalized to the range [0,1] by the ratio of the deviation between the current actual completion time and the planned completion time to the total planned time; the real-time fault status or fault probability of the key equipment is obtained based on the equipment's real-time health monitoring data, predictive maintenance model, and fault indication; the real-time utilization rate of the key resources is calculated by the ratio of the current occupancy rate to the maximum capacity; and the environmental risk factor is based on the impact of external environmental factors such as real-time weather, traffic, and emergencies on the process. The degree of influence is calculated and normalized to the range of [0,1]; the set weights are obtained by training based on expert experience and historical data.

[0092] For example, if a ship's ETA is delayed by 2 hours and the planned total sailing time is 24 hours, then .

[0093] If the port crane malfunctions, then If the equipment health status alert is issued, then It is a value between [0,1].

[0094] In one feasible embodiment of the present invention, the downstream link Real-time capacity and buffering capacity assessment values The calculation formula is: ,in Indicates downstream links The remaining processing capacity, Indicates downstream links The planned relaxation time, Indicates downstream links The availability of backup resources, Indicates downstream links Demand pressure These are the weights corresponding to the set remaining processing capacity, planned slack time, availability of backup resources, and demand pressure, respectively. .

[0095] It should be noted that the remaining processing capacity is calculated as a percentage of the current available processing capacity to the total capacity; the planned slack time is calculated as a percentage of the elastic time or buffer time reserved in the subsequent plan to the total task time; the availability of backup resources indicates whether there are available backup equipment, backup manpower, or alternative paths; and the demand pressure is calculated based on the ratio of the current order volume, the predicted throughput value, and the average value, and normalized to the range of [0,1].

[0096] S4. Coupling Risk Identification and Assessment: Based on the real-time coupling coefficient, a dynamic risk map of the multimodal transport route is constructed in the multimodal transport digital twin, and the real-time coupling coefficients of several adjacent links of the multimodal transport route are aggregated to calculate the risk indicator system.

[0097] In one feasible embodiment of the present invention, the specific construction process of the dynamic risk map of the multimodal transport route includes: rendering the topology of the multimodal transport route in the multimodal transport digital twin on an interactive visual monitoring screen in real time, wherein the color of each link is dynamically gradient rendered according to its corresponding real-time coupling coefficient, and when the real-time coupling coefficient exceeds a preset risk threshold, it is highlighted and an alarm is triggered, thereby constructing the dynamic risk map of the multimodal transport route.

[0098] It should be noted that the color gradient rendering specifically includes: dynamically rendering the color of each link according to its corresponding real-time coupling coefficient. The colors employ a continuous, perceptibly uniform gradient color spectrum. For example, a smooth transition from green to yellow, and then to red. This color mapping algorithm is implemented based on a predefined threshold range and a color interpolation function.

[0099] For example, when 0≤ The value is green when it is less than 0.3, and green when it is 0.3 or less. Yellow is indicated when <0.7, and yellow is indicated when 0.7≤ When the value is ≤1, it is red, and linear or nonlinear interpolation is performed within this interval.

[0100] It should be explained that the green color indicates low coupling risk. Approaching 0; the yellow color indicates a moderate coupling risk. Approximately 0.5; the red color indicates a high risk of coupling. Close to 1.

[0101] The highlighted display and warning specifically include: when any When the value exceeds a preset risk threshold, a risk alarm is automatically triggered. The alarm is triggered through various means, such as highlighting a corresponding link on a visual interface (e.g., flashing red and bold lines), playing an alarm sound, or sending a notification via email, SMS, or instant messaging. The risk threshold can be dynamically configured based on business needs, cargo value, and risk tolerance.

[0102] In one feasible embodiment of the present invention, the risk indicator system includes, but is not limited to, the overall risk exposure of multimodal transport routes, the risk contribution of key links, and the maximum risk propagation chain.

[0103] By aggregating the real-time coupling coefficients of several adjacent links in a multimodal transport route, and calculating the overall risk exposure, risk contribution of key links, and maximum risk propagation chain of the multimodal transport route according to standard calculation formulas, the risk exposure of the multimodal transport route, the risk contribution of key links, and the maximum risk propagation chain are obtained respectively.

[0104] It should be noted that the overall risk exposure of the described path Specifically, it measures the total coupling risk intensity of the entire multimodal transport route, reflecting the overall coupling risk level faced by the route, with a value range of [0, N], where N is the total number of link pairs. The specific calculation steps are as follows: (1) Determine the range of link pairs: Traverse all link pairs with edges in the directed graph model of the multimodal transport digital twin. Excluding logically unrelated link pairs, such as Shanghai highway and Guangzhou shipping, we obtain the total number of valid link pairs M, and then calculate the risk weight of each link pair. The values ​​of 0.6 and 0.4 represent the weights of expert experience and can be adjusted according to the scenario.

[0105] The For the link The throughput capacity weight is determined by the process. The ratio of the current load rate to the maximum throughput capacity is calculated.

[0106] For example, if the current load factor of Shanghai Port is 80% and the maximum throughput capacity is 100 containers / hour, then at this time... =0.8.

[0107] The For the link Time sensitivity weight, through process The ratio of the planned time margin to the standard transportation time is calculated.

[0108] For example, if the planned time margin for the highway transport segment is 30 minutes and the standard transport time is 120 minutes, then at this time... =0.25.

[0109] (2) Calculate the overall risk exposure: .

[0110] For example, a certain multimodal transport route has three effective links. Thus, through calculation, we can obtain .

[0111] (3) Risk level classification: based on Values ​​are categorized into levels.

[0112] For example, at a low risk level, <0.5, 0.5≤ for medium risk level <1.2, high-risk level ≥1.2.

[0113] Risk contribution of the key links Specifically, it measures the contribution of a single link to the overall path coupling risk, identifies the core source of risk, and takes values ​​in the range [0,1]. The higher the value, the more critical the risk source. The calculation steps are as follows: (1) Construct a risk contribution matrix: with links as rows or columns, the matrix elements are... Representation Steps pass For the process The risk contribution, and ,in That is, the process Risk percentage across all related processes.

[0114] (2) Introduce the PageRank algorithm to calculate the initial contribution: Treat each link as a graph node, Considering the transition probabilities between nodes, the initial risk contribution of each stage is calculated iteratively using PageRank. The iterative formula is: ,in Here, 0.85 is the number of iterations, 0.85 is the damping coefficient, and N is the total number of elements. The iteration continues until... Stop when the time comes.

[0115] (3) Correcting contribution: Correcting the initial contribution based on the historical risk records of each stage. value: ,in For the link The historical risk coefficient, and .

[0116] (4) Key link identification: according to Sort the steps in descending order, select the top 20% as key risk steps, and mark their main contribution paths, such as step A. Mainly from B and C.

[0117] The largest risk propagation chain Specifically, it identifies the coupled risk chain with the highest risk transmission intensity and the most affected links in the entire path, reflecting the maximum potential impact range of the risk. The specific calculation steps are as follows: (1) Screening high-risk initiating links: First, determine the candidate set of risk sources and select ≥0.15 key links or The source link is ≥0.8.

[0118] (2) Risk propagation path search: based on candidate risk sources Starting from the edge of the directed graph, as shown below... Search, keep only For high-risk edges with a risk level ≥ 0.4, calculate the path risk intensity for each searched path. It is all in the path The product of these. And record the number of affected links for each path, such as... It affected three aspects.

[0119] For example, suppose of If they are 0.7 and 0.6 respectively, then .

[0120] (3) Determine the chain of greatest risk propagation: prioritize sorting by path risk intensity in descending order. The highest-ranking path is the candidate chain; if there are multiple paths... Similarly, choose the path that affects the most links.

[0121] For example, path 1 is 3 steps, path 2 is If there are two links, then the path with the greatest risk of transmission is path 2.

[0122] (4) Risk assessment of the propagation chain: Calculate the terminal risk intensity of the maximum risk propagation chain, i.e., the cumulative risk of the last link in the chain. If the terminal risk intensity is ≥0.6, it is considered a high-risk transmission chain and requires priority handling.

[0123] This invention provides multi-dimensional and quantitative risk assessment results by evaluating the coupling risks between various links in a multimodal transport route in real time, dynamically and quantitatively. This makes risk management more refined and scientific, improves the resilience and efficiency of multimodal transport operations, and reduces potential economic losses and operational interruption risks.

[0124] S5. Risk Visualization: Based on the multimodal transport digital twin, real-time coupling coefficient and risk indicator system, a multimodal transport route environment coupling risk identification report is generated. When the risk indicators exceed the preset threshold, multi-level early warning information is generated and visualized.

[0125] In one feasible embodiment of the present invention, the specific content of step S5 includes: generating a multimodal transport route environment coupling risk identification report specifically involves: organizing the risk report content of multimodal transport, listing in detail all currently identified potential risk sources, visually displaying key risk propagation paths, and indicating the probability and propagation time of each transport link of multimodal transport being affected by risks.

[0126] It should be noted that the risk report includes the scope and time of risk identification, the overall risk assessment results, and the risk level.

[0127] All potential risk sources include their type, location, initial risk probability, and potential scope of impact.

[0128] The generation of multi-level early warning information specifically includes: the early warning information includes the early warning level, early warning type, affected links and paths, and suggested actions.

[0129] The warning levels are divided into different levels based on the urgency, severity, and scope of the risk. The recommended actions include activating backup transportation plans, suspending subsequent operations, or notifying relevant carriers to adjust their plans.

[0130] This invention continuously receives real-time data and dynamically updates risk assessment results. When risk indicators exceed preset thresholds, it promptly triggers early warnings, thereby enhancing the risk warning capabilities of multimodal transport routes. This helps managers respond quickly and take countermeasures, effectively reducing the probability and impact of risk events.

[0131] Example 2

[0132] Please see Figure 2 As shown, the present invention provides a multimodal transport route link coupling risk identification system, the specific module distribution is as follows: data acquisition module, digital twin construction module, dynamic coupling coefficient calculation module, coupling risk identification and assessment module, and risk visualization display module.

[0133] The data acquisition module is connected to the digital twin construction module, the digital twin construction module is connected to the dynamic coupling coefficient calculation module, the dynamic coupling coefficient calculation module is connected to the coupling risk identification and assessment module, and the coupling risk identification and assessment module is connected to the risk visualization display module.

[0134] The data acquisition module collects and preprocesses process data and environmental perception data of multimodal transport routes through IoT sensing devices.

[0135] The digital twin construction module maps process data and environmental perception data to corresponding nodes in the constructed multimodal transport path environment network model, thereby constructing a multimodal transport digital twin.

[0136] The dynamic coupling coefficient calculation module, based on the multimodal transport digital twin, defines and calculates the real-time coupling coefficient between each pair of adjacent links in the multimodal transport path.

[0137] The coupling risk identification and assessment module constructs a dynamic risk map of multimodal transport routes in the multimodal transport digital twin based on real-time coupling coefficients, and aggregates the real-time coupling coefficients of several adjacent links of the multimodal transport route to calculate the risk indicator system.

[0138] The risk visualization module generates a multimodal transport route environment coupling risk identification report based on the multimodal transport digital twin, real-time coupling coefficient, and risk indicator system. When the risk indicators exceed the preset threshold, it generates multi-level early warning information and displays it visually.

[0139] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0140] Those skilled in the art will recognize that the algorithmic steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0141] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0142] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0143] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for quantitative assessment and early warning of coupling risks in multimodal transport routes, characterized in that: include: S1. Data Acquisition: Collect and preprocess process data and environmental perception data of multimodal transport routes through IoT sensing devices; S2. Construction of digital twin: Map process data and environmental perception data to the corresponding nodes in the constructed multimodal transport path environment network model to construct a multimodal transport digital twin; S3. Dynamic Coupling Coefficient Calculation: Based on the multimodal transport digital twin, define and calculate the real-time coupling coefficient between each pair of adjacent links in the multimodal transport path. S4. Coupling Risk Identification and Assessment: Based on the real-time coupling coefficient, a dynamic risk map of the multimodal transport path is constructed in the multimodal transport digital twin, and the real-time coupling coefficients of several adjacent links of the multimodal transport path are aggregated to calculate the risk indicator system. S5. Risk Visualization: Based on the multimodal transport digital twin, real-time coupling coefficient and risk indicator system, a multimodal transport route environment coupling risk identification report is generated. When the risk indicators exceed the preset threshold, multi-level early warning information is generated and visualized.

2. The method for quantitative assessment and early warning of coupling risks in multimodal transport routes according to claim 1, characterized in that: The specific construction process of the multimodal transport route environment network model includes: Each independent transportation link in the multimodal transport route is defined as a node, and each node is assigned a link type, geographical coordinates, planned duration, carrier name, historical average risk level, maximum throughput capacity, and current status feature vector. The connection between two adjacent or logically related links is defined as an edge, and the edge is assigned weights such as logistics cost, time cost, initial value of risk propagation probability, and actual distance. Based on the defined nodes and edges, a directed graph model is constructed, and the directed graph model is stored in the form of an adjacency matrix or an adjacency list, thereby building a network model for multimodal transport path environment.

3. The method for quantitative assessment and early warning of coupling risks in multimodal transport routes according to claim 2, characterized in that: The specific construction process of the multimodal transport digital twin includes: The multimodal transport route environment network model is placed in the digital space, and the link data and environmental perception data of the multimodal transport route are mapped to the multimodal transport route environment network model in the digital space to achieve accurate mapping of the physical world, thereby constructing a multimodal transport digital twin. Inject a series of physical rules and business logic into each node, edge, and their interactions in the digital twin; The physical rules include kinematic rules, capacity constraint rules, equipment performance rules, and environmental impact rules, while the business logic includes process dependency logic, resource scheduling logic, timeliness constraint logic, and exception handling logic.

4. The method for quantitative assessment and early warning of coupling risks in multimodal transport routes according to claim 1, characterized in that: Define the expression for each pair of adjacent links in the multimodal transport route as follows: The expression for the real-time coupling coefficient is defined as follows: ; The real-time coupling coefficient It is a dynamic, dimensionless value between [0,1] used to quantify upstream processes. Potential risks or abnormal conditions can be transmitted to downstream processes. and downstream links The likelihood and intensity of negative impacts on normal operation; The real-time coupling coefficient The specific calculation formula is as follows: ,in Indicates upstream link Real-time risk status assessment value, Indicates downstream links Real-time capacity and buffering capacity assessment values, Representation Steps With links The inherent coupling strength between them It is a non-linear mapping function.

5. The method for quantitative assessment and early warning of coupling risks in multimodal transport routes according to claim 4, characterized in that: The upstream link Real-time risk status assessment value The calculation formula is ,in Indicates upstream link Time deviation rate Upstream links Real-time fault status or fault probability of critical equipment. Indicates upstream link Real-time utilization rate of critical resources Indicates upstream link Environmental risk factors These are the set weights for time deviation rate, failure probability, real-time utilization rate, and environmental risk factor, respectively. .

6. The method for quantitative assessment and early warning of coupling risks in multimodal transport routes according to claim 4, characterized in that: The downstream links Real-time capacity and buffering capacity assessment values The calculation formula is: ,in Indicates downstream links The remaining processing capacity, Indicates downstream links The planned relaxation time, Indicates downstream links The availability of backup resources, Indicates downstream links Demand pressure These are the weights corresponding to the set remaining processing capacity, planned slack time, availability of backup resources, and demand pressure, respectively. .

7. The method for quantitative assessment and early warning of coupling risks in multimodal transport routes according to claim 1, characterized in that: The specific process of constructing the dynamic risk map of the multimodal transport route includes: The multimodal transport route topology in the multimodal transport digital twin is rendered in real time on an interactive visual monitoring screen. The color of each link is dynamically gradient-rendered according to its corresponding real-time coupling coefficient. When the real-time coupling coefficient exceeds a preset risk threshold, it is highlighted and an alarm is triggered. Based on this, a dynamic risk map of the multimodal transport route can be constructed.

8. The method for quantitative assessment and early warning of coupling risks in multimodal transport routes according to claim 1, characterized in that: The risk indicator system includes, but is not limited to, the overall risk exposure of multimodal transport routes, the risk contribution of key links, and the maximum risk propagation chain; By aggregating the real-time coupling coefficients of several adjacent links in a multimodal transport route, and calculating the overall risk exposure, risk contribution of key links, and maximum risk propagation chain of the multimodal transport route according to standard calculation formulas, the risk exposure of the multimodal transport route, the risk contribution of key links, and the maximum risk propagation chain are obtained respectively.

9. The method for quantitative assessment and early warning of coupling risks in multimodal transport routes according to claim 1, characterized in that: The specific content of step S5 includes: The generated multimodal transport route environment coupling risk identification report specifically includes: organizing the risk report content of multimodal transport, listing in detail all currently identified potential risk sources, visually demonstrating key risk propagation paths, and indicating the probability of each transport link in multimodal transport being affected by risks and their propagation time; The generation of multi-level early warning information specifically includes: the early warning information includes the early warning level, early warning type, affected links and paths, and suggested actions; The warning levels are divided into different levels based on the urgency, severity, and scope of the risk. The recommended actions include activating backup transportation plans, suspending subsequent operations, or notifying relevant carriers to adjust their plans.

10. A system for implementing a quantitative assessment and early warning method for the coupling risk of multimodal transport routes as described in claims 1-9, characterized in that: include: The data acquisition module collects and preprocesses process data and environmental perception data of the multimodal transport route through IoT sensing devices; The digital twin construction module maps process data and environmental perception data to corresponding nodes in the constructed multimodal transport path environment network model, thereby constructing a multimodal transport digital twin. The dynamic coupling coefficient calculation module, based on the multimodal transport digital twin, defines and calculates the real-time coupling coefficient between each pair of adjacent links in the multimodal transport path. The coupling risk identification and assessment module constructs a dynamic risk map of multimodal transport routes in the multimodal transport digital twin based on real-time coupling coefficients, and aggregates the real-time coupling coefficients of several adjacent links of the multimodal transport route to calculate the risk indicator system. The risk visualization module generates a multimodal transport route environment coupling risk identification report based on the multimodal transport digital twin, real-time coupling coefficient, and risk indicator system. When the risk indicators exceed the preset threshold, it generates multi-level early warning information and displays it visually.

Images