Multi-dimensional data fusion and intelligent decision-making-based real-time dynamic scheduling system for hazardous chemical vehicles in chemical industry park

Abstract

A chemical industrial park hazardous chemical substance vehicle real-time dynamic scheduling system based on multi-dimensional data fusion and intelligent decision-making comprises a vehicle monitoring module used for collecting dynamic information of transport vehicles in real time and monitoring the running state of each vehicle; the risk assessment module is used for calculating the current risk level of the transportation task based on the data collected in real time in combination with a risk assessment model; the path optimization module is used for calculating and adjusting an optimal driving path through a dynamic A * path planning algorithm; and the emergency response module is used for automatically starting when the high-risk condition is detected, so that the safety in the dangerous chemical transportation process is ensured. The invention aims to realize real-time and dynamic scheduling of the hazardous chemical vehicles in the chemical industrial park through multi-dimensional data fusion and intelligent algorithm design. By integrating multi-dimensional data of vehicles, goods, external environment, time, space and the like, the system can be ensured to automatically adjust a scheduling scheme in an emergency, and an efficient emergency response mechanism is provided.

Description

Technical Field

[0001] The invention relates to a real-time dynamic dispatching system for hazardous chemical vehicles in a chemical park based on multi-dimensional data fusion and intelligent decision-making. Background Art

[0002] The transportation of hazardous chemicals in chemical parks poses extremely high safety challenges. Hazardous chemicals are flammable, explosive, and toxic. Once an accident occurs, it often has a serious impact and may cause casualties and environmental pollution. During transportation, multiple factors need to be considered comprehensively, such as vehicle status, cargo characteristics, road conditions, external environment (such as weather conditions), spatial area restrictions, and time restrictions, to ensure safe arrival at the destination.

[0003] The management of hazardous chemicals transportation is very complex, mainly reflected in: data acquisition and processing, real-time monitoring, risk assessment and prevention, emergency response, etc. Traditional manual management and simple GPS positioning can no longer meet complex management needs, and more intelligent and comprehensive management methods are needed to cope with changing transportation environments and risk events.

[0004] The existing management methods for the transportation of hazardous chemicals have the following deficiencies:

[0005] 1. Insufficient data support

[0006] The existing hazardous chemicals transportation dispatch system mainly relies on simple GPS positioning and basic route planning, lacking full utilization and intelligent analysis of multi-dimensional data. This lack of data support makes it difficult for the system to fully evaluate various risk factors in the transportation process, such as changes in the external environment and changes in the status of goods, thus affecting the accuracy and timeliness of dispatch decisions.

[0007] 2. Delayed emergency response

[0008] Traditional dispatch systems have a slow response speed and lack of immediate emergency response capabilities when facing emergencies (such as traffic accidents and bad weather). This lag may lead to the expansion of accidents and aggravation of consequences, affecting the overall transportation safety and efficiency.

[0009] 3. Insufficient intelligent dispatching

[0010] The existing system lacks the ability to intelligently analyze and dynamically adjust real-time data, and most decisions rely on static plans and simple forecasting models. This deficiency makes the system limited in its ability to deal with complex transportation environments and changing situations, and easily leads to untimely or inaccurate scheduling decisions. Summary of the invention

[0011] The present invention aims to achieve real-time and dynamic dispatch of hazardous chemical vehicles in chemical parks through multi-dimensional data fusion and intelligent algorithm design. By integrating multi-dimensional data such as vehicles, goods, external environment, time, and space, the system can automatically adjust the dispatch plan in emergency situations and provide an efficient emergency response mechanism.

[0012] A real-time dynamic dispatching system for hazardous chemicals vehicles in chemical parks based on multi-dimensional data fusion and intelligent decision-making, including:

[0013] The vehicle monitoring module is used to collect dynamic information of transport vehicles in real time, monitor the operating status of each vehicle, and process the data collected by sensors;

[0014] The risk assessment module is used to calculate the current risk level of the transportation task based on the real-time collected data combined with the risk assessment model;

[0015] Path optimization module, which is used to calculate and adjust the optimal driving route based on real-time traffic data, environmental conditions and mission requirements through the dynamic A* path planning algorithm;

[0016] The emergency response module is used to automatically start when a high-risk situation is detected to ensure the safety of hazardous chemicals during transportation.

[0017] The present invention provides a real-time dynamic dispatching system for hazardous chemicals vehicles in chemical parks based on multi-dimensional data fusion and intelligent decision-making. By introducing multi-dimensional data fusion and intelligent decision-making systems, the deficiencies of traditional hazardous chemicals transportation dispatching systems in data support, emergency response and intelligent decision-making are made up. Through real-time monitoring and intelligent analysis, the system can effectively improve the safety, efficiency and management cost optimization of transportation, bringing new intelligent solutions to hazardous chemicals transportation management. DETAILED DESCRIPTION

[0018] The core of the present invention is to integrate multi-dimensional data in the process of hazardous chemicals transportation through the Internet of Things technology, including real-time monitoring of vehicles, goods, external environment, time and space constraints and other data.

[0019] 1. Multi-dimensional data fusion and real-time monitoring

[0020] (1) Vehicle dynamic data: Vehicle speed, location, load, fuel consumption, braking status and other data are collected in real time using on-board sensors, GPS and other technologies. The system dynamically adjusts scheduling decisions based on the vehicle status to ensure that the vehicle is in the best operating condition.

[0021] (2) Cargo monitoring data: The system obtains the status data of hazardous chemicals (such as temperature, pressure, concentration, etc.) through the on-board monitoring equipment. If an abnormal cargo status is detected (for example, the temperature or pressure exceeds the safe range), the system will immediately trigger an alarm and activate the emergency mechanism.

[0022] (3) External environmental data: By connecting to weather monitoring systems, traffic monitoring systems, etc., the system can obtain real-time weather data (such as heavy rain, high temperature, wind speed, etc.) and traffic conditions (such as accidents, congestion, etc.). These external data help the system adjust the vehicle's route in real time.

[0023] (4) Transportation space area restriction data: The system divides risk areas inside and outside the park through GIS (geographic information system) to ensure that vehicle routes avoid high-risk areas (such as explosive warehouses, densely populated areas, etc.).

[0024] (5) Transportation time limit data: The system ensures that vehicles are safely transported within the permitted time period based on the actual production tasks of the park and the time limit for the transportation of hazardous chemicals stipulated by the park. The system also reasonably arranges the transportation time according to the urgency and timeliness of the task to ensure efficient completion of the task.

[0025] 2. Dynamic scheduling mechanism based on risk classification

[0026] (1) Model background

[0027] In order to more accurately evaluate the various risks in hazardous chemicals transportation tasks, the present invention designs a dynamic weighted risk assessment model. Taking into account the complex interaction between multidimensional data, a nonlinear weighting function is adopted, and cross factors and dynamic adjustment factor mechanisms are introduced to more accurately reflect the risks of tasks under different conditions.

[0028] (2) Risk Assessment Factors

[0029] The assessment model comprehensively considers the following multi-dimensional risk factors:

[0030] Cargo Hazard Rating (H): The hazardous chemicals are scored based on their category, characteristics, storage temperature, pressure, etc. Including but not limited to flammable, explosive, and toxic substances.

[0031] Vehicle Status Score (V): A comprehensive assessment of the vehicle's current operating status, including health status, maintenance records, real-time monitoring data, etc., especially vehicle failure rate, driver experience, vehicle load, etc.

[0032] Environmental risk score (E): Based on the assessment of external environmental factors such as real-time weather conditions, traffic flow, road conditions, air quality, etc. Bad weather, congested roads, polluted air quality, etc. will increase this score.

[0033] Spatial risk score (S): The vehicle is scored based on whether its route passes through high-risk areas (such as tank areas, chemical storage areas, densely populated areas, accident-prone areas, etc.). The risk score is higher if it passes through these areas.

[0034] Time risk score (T): Based on the time urgency of the transport task, whether it is during peak hours or special times (such as night transport, holidays). The score increases when time is urgent or the transport is carried out during special periods.

[0035] (3) Risk scoring formula

[0036] In order to better adapt to the complexity of different tasks, we use a nonlinear weighting function and introduce a cross factor to reflect the mutual influence between different risk factors. In addition, since the specific conditions of the task (such as weather, vehicle condition, cargo properties, etc.) will change dynamically, the weight of each factor, α, β, γ, δ,∈, is dynamically adjusted and can be updated at any time.

[0037] The risk score formula is:

[0038]

[0039] 1) Normalization factor

[0040]

[0041] Normalization is to unify each factor to the range of 0-1 to solve the problem of different magnitudes between factors. For example, the cargo hazard score H may be 1-10, while the time risk score T may be 1-100. Without normalization, weight adjustment will become difficult and the score results will be unbalanced. Normalization ensures that each risk factor in the formula is weighted at the same magnitude, thereby avoiding excessive influence of a certain factor on the overall score.

[0042] 2) Linear terms and logarithmic terms

[0043]

[0044] Linear terms (such as α1×H 归一化 ): Indicates the direct impact of each factor on the risk score. The role of the linear term is to reflect the importance of each factor under normal circumstances.

[0045] Logarithmic terms (such as ): Used to deal with the risk amplification effect in extreme cases. The growth rate of the logarithmic term is fast at first and then slows down, which can significantly improve the risk score when the factor value is small, but when the factor value is high, the amplification effect gradually weakens.

[0046] The logarithmic term supplements the shortcomings of the linear model for nonlinear risks, especially in the case of dangerous goods (such as flammable and explosive goods) or harsh environments (such as blizzards), and can more accurately reflect the risk level.

[0047] 3) Index weighting of spatial risk

[0048]

[0049] The spatial risk score S reflects the risk of a vehicle passing through a specific area (such as a storage tank area or a densely populated area). An exponential form is used to describe the amplification effect of this risk on the overall score.

[0050] When S is small, the impact of exponential weighting on the overall score is weak; but when S is large (such as vehicles passing through the core area of ​​a chemical park), exponential weighting will quickly amplify the overall risk score.

[0051] Through the index form, the system can more sensitively capture the impact of high-risk areas, prioritize adjusting driving routes or increasing monitoring frequency.

[0052] 4) Linear and nonlinear terms of time risk

[0053] ∈1×T 归一化 +∈2×T 2

[0054] Linear term (∈1×T 归一化 ): Describes the direct impact of time risk on the overall score. For example, the urgency of a transportation task during peak hours will significantly increase the risk score.

[0055] Nonlinear term (∈2×T 2 ): When the time risk score T exceeds the normal range (such as extremely urgent tasks or night transportation), the square term will quickly amplify the impact of the time factor on the overall score.

[0056] The system can more accurately distinguish between general tasks and urgent tasks, and automatically adjust scheduling priorities based on the time requirements of the tasks.

[0057] 5) Cross-term

[0058] η1×T×E+η2×H×S

[0059] Time and environment interaction term (η1×T×E): reflects the combined effect of time pressure and environmental risk. For example, if bad weather occurs during peak hours, the combined effect of time risk and environmental risk will significantly increase the overall risk score.

[0060] The interaction term between cargo and space (η2×H×S) reflects the amplification effect of dangerous goods transported through high-risk areas. For example, when flammable and explosive goods are transported near densely populated areas, their danger will be further increased.

[0061] The cross terms capture the interactions between the factors, more carefully reflecting the comprehensive impact of multidimensional risk factors on the overall score and improving the accuracy of the model.

[0062] 6) Comprehensive explanation

[0063] The entire formula combines linear terms, nonlinear terms, exponential weighting of spatial risks, and the cross-effects of time and environment, cargo and space, forming a complex but accurate risk assessment mechanism.

[0064] Basic scoring part:

[0065] α, β, and γ represent the direct risk scores of cargo, vehicles, and environment.

[0066] H 2 V 2 E 2 Through logarithmic function correction, we can avoid the score from growing too fast and adapt to extreme risks.

[0067] Space risk amplification:

[0068] This reflects the significant amplifying effect of high-risk areas on the overall score.

[0069] Time impact:

[0070] ∈1×T 归一化 +∈2×T 2 Reflects the direct and nonlinear impact of time urgency on the overall score.

[0071] Interactions between factors:

[0072] η1×T×E+η2×H×S describes the compound risk under the interaction of multidimensional factors, which enhances the complexity and flexibility of the formula.

[0073] (4) Risk level classification

[0074] Based on the improved risk score R, the system will evaluate each transport task in real time, determine its risk level, and trigger the corresponding scheduling strategy. The evaluation results can be divided into three risk levels:

[0075] Low risk (R<30): The task can be executed along the normal route without additional intervention. The system will monitor real-time data, but no special adjustments to the path are required.

[0076] Medium risk (30≤R<60): At this time, the task needs to increase the monitoring frequency, and the system will dynamically adjust the path to avoid potential high-risk areas. At the same time, the system will prepare backup plans to deal with possible emergencies.

[0077] High risk (R≥60): The system triggers the emergency plan, replans the route, prioritizes avoiding high-risk areas, and notifies relevant emergency departments to ensure the safe completion of the mission. At this time, all safety resources (such as emergency convoys, special police, firefighters, etc.) will be prepared in advance.

[0078] (5) Dynamic adjustment factor

[0079] In the system of the present invention, the weight of each risk factor α, β, γ, δ, ∈ will be adjusted in real time according to the characteristics of the transportation task. For example:

[0080] Changes in the transportation environment: In severe weather or special road conditions, the system will automatically increase the weight γ of the environmental risk score.

[0081] Vehicle failure rate: When the vehicle failure rate is high, the system will increase the weight β of the vehicle status score to make it have a greater impact on the final risk score.

[0082] (6) Real-time feedback and adjustment of the system

[0083] During the execution of a task, the system will monitor the running status of the task in real time and dynamically adjust the risk score and scheduling strategy based on real-time data feedback. For example, when the external environment or vehicle status changes, the system will automatically adjust the weights in the risk score formula and update the scheduling strategy in real time to ensure that the task is always within a controllable safety range.

[0084] (7) Summary

[0085] The risk assessment model of the present invention can more accurately reflect the risks in complex transportation tasks by introducing nonlinear weighting functions and dynamic adjustment factors. Compared with traditional linear models, this model can more carefully evaluate the interactions between various factors, especially when multiple factors are intertwined (such as cargo status, weather, vehicle status, road conditions, and time constraints, etc.), which greatly improves the accuracy and timeliness of risk assessment. Through this model, the system can monitor and optimize the safety of hazardous chemicals transportation tasks in real time, providing more reliable decision support for the scheduling system.

[0086] 3. Intelligent path optimization and scheduling adjustment

[0087] The system uses a dynamic A* path planning algorithm and combines real-time data to make scheduling adjustments to ensure that the vehicle travels on the optimal path. The specific steps are as follows:

[0088] (1) Path optimization algorithm steps:

[0089] Initialize path planning: The system initializes the A* path planning algorithm based on the vehicle's departure and destination, as well as the current real-time environmental data, and calculates the initial optimal path.

[0090] The execution process is:

[0091] The system first obtains the current location (departure point) and the designated destination coordinates through the vehicle's GPS, and combines it with the GIS system to obtain map data of all possible routes.

[0092] Initialize the A* algorithm, set the starting node n_start and the target node n_goal, divide the path search space into multiple small nodes, and perform a preliminary search on all path nodes.

[0093] Starting from the initial node, the calculation begins according to the principle of shortest distance, and potential obstacles, traffic conditions, etc. on the path are evaluated.

[0094] Weight setting: Assign weights based on factors such as road conditions, traffic flow, external environment, space and time. Assign dynamic weights to each path node based on different factors. Road sections with lower weights are given priority.

[0095] The execution process is:

[0096] The system sets an initial weight for each path node and dynamically adjusts the weight based on factors such as road conditions, traffic flow, external environment (such as weather conditions), regional risks, and time requirements.

[0097] Road condition factors: The system uses real-time traffic data to determine the degree of road congestion, whether there are traffic accidents, construction areas, etc. For these influencing factors, set a higher weight value and give priority to avoiding them.

[0098] Spatial factors: If a vehicle needs to pass through a high-risk area (such as a hazardous goods storage area or a densely populated area), the system will increase the weight of the area to avoid unnecessary high-risk areas.

[0099] Time requirement: The system analyzes the urgency of the transport task. If the task needs to be completed as soon as possible, the system will give priority to the path with low time cost and reduce the weight on certain paths.

[0100] Heuristic function (h): The algorithm uses the heuristic function h(n) to estimate the cost from node n to the target node, taking into account the length of the road section, traffic conditions, risk areas, etc. The cost of the path nodes is evaluated through the heuristic function to ensure that the optimal path is selected.

[0101] The execution process is:

[0102] In the A* algorithm, the heuristic function h(n) is used to estimate the cost from node n to the target node. Its calculation method combines the path length, road conditions, and risk assessment results.

[0103] The calculation formula of the heuristic function is:

[0104] h(n)=d(n,n_goal)+∑risk(n) (2)

[0105] Where d(n,n_goal) is the distance from node n to the target node n_goal, and ∑risk(n) is the cumulative risk assessment value of the current path node.

[0106] Risk assessment not only considers the length of the road section, but also combines comprehensive factors such as spatial risk and temporal risk.

[0107] Path search and evaluation: The algorithm expands the preferred path nodes, selects the optimal path, and updates the path weights based on real-time data to ensure that the vehicle travels on the safest route.

[0108] The execution process is:

[0109] The algorithm starts from the starting node n_start and gradually expands the path along the node with the smallest weight. Each time it expands, the algorithm will select the next node with the lowest path cost and give priority to exploring a better path.

[0110] The system monitors the weight updates of all nodes in real time. If the risk assessment score of a path exceeds the set threshold (for example, a traffic accident occurs on the road or the weather deteriorates), the algorithm will exclude it from the optimal path candidate.

[0111] Continuously evaluate the security, time cost, and external environmental impact of the expansion path to ensure that the path remains the optimal choice during the expansion process.

[0112] Dynamic adjustment: When the system detects changes in the external environment (such as sudden accidents or worsening weather), the algorithm will automatically replan the route to avoid the affected areas.

[0113] The execution process is:

[0114] The system updates the data of all path nodes every second to assess whether there are sudden changes, such as road congestion, traffic accidents or sudden deterioration of weather.

[0115] Once an abnormality in the environment or vehicle status is detected (such as abnormal status of hazardous materials), the algorithm will recalculate and update the path in real time to ensure that the vehicle can adjust to a safe driving route in the shortest time.

[0116] (2) Scheduling adjustment mechanism steps:

[0117] Real-time data monitoring: The system collects real-time data on vehicles, cargo, external environment, space and time every second, and updates the evaluation model based on the data.

[0118] The execution process is:

[0119] The system collects real-time data provided by vehicle sensors every second (such as speed, location, fuel consumption, and cargo status), and integrates it with external environmental data (such as weather and road conditions), spatial data, and temporal data through the Internet of Things platform.

[0120] Based on real-time data, the system will continuously update the current status of the vehicle and compare it with the expected data of the evaluation model. If the data is abnormal (such as a sudden slowdown in vehicle speed, a rise in cargo temperature, etc.), the system will mark the task as a potential risk.

[0121] Detect risk factors: Identify and evaluate risk factors that affect vehicle transportation. If the system finds that the risk score rises to medium or high risk, it will immediately enter the scheduling adjustment process.

[0122] The execution process is:

[0123] The system inputs multidimensional data into the assessment model and calculates the real-time risk score R. When the score exceeds the set threshold (such as 30 or 60 points), the system will enter the scheduling adjustment mode.

[0124] Through the calculation of the aforementioned risk assessment model, the current risk level of the task is comprehensively calculated by combining multiple scores such as the dangerousness of the cargo, the vehicle status, the external environment, space constraints, and time constraints.

[0125] The threshold of the risk assessment score is adjusted dynamically to ensure that potential problems can be discovered in advance in special emergency situations.

[0126] Path replanning: When the risk score increases, the system quickly adjusts the path planning to ensure safety. The system recalculates the optimal path through the dynamic A* algorithm and notifies the driver of the new route.

[0127] The execution process is:

[0128] When the system detects that a mission is at medium or high risk (score ≥ 30), it immediately initiates a dynamic A* path recalculation. Based on the latest real-time data, the system regenerates a route that avoids high-risk areas or traffic congestion.

[0129] After the new path is generated, the system pushes the updated route to the driver through the vehicle system or mobile application, and sends risk alerts to the dispatcher to ensure that the driver and dispatcher can respond quickly when the task status changes.

[0130] Emergency response trigger: In high-risk conditions, the emergency plan is activated to ensure transportation safety. If the risk score reaches high risk, the system triggers the emergency plan, notifies the emergency department and guides the vehicle to a safe area.

[0131] The execution process is:

[0132] When the system detects that the risk score R reaches a high risk (for example, ≥60), the system will immediately trigger the emergency response mechanism. First, the system will notify the nearest emergency command center and relevant emergency departments to make safety preparations in advance.

[0133] At the same time, the system will plan a route for the vehicle to the nearest safe zone and notify the driver of the latest route through the mobile terminal. The emergency response process also includes emergency monitoring of the transported goods. If necessary, the system will require the driver to stop midway to prevent further harm.

[0134] In extreme cases (such as vehicle breakdown or cargo leakage), the system will also arrange a special emergency fleet or rescue force for rapid response.

[0135] 4. Improve scheduling efficiency

[0136] 4.1 Millisecond-level path planning

[0137] The intelligent dispatching system uses the dynamic A* path planning algorithm and combines multi-dimensional data for efficient task allocation and path calculation. The algorithm uses real-time updated data to perform fast calculations during path search, evaluates various factors on the path (such as vehicle status, cargo status, traffic, weather, space, and time constraints, etc.), and ensures that the calculation results can quickly adapt to changes. Because the algorithm uses heuristic methods during the search process, it can quickly screen out priority paths, greatly reducing unnecessary calculation overhead.

[0138] Millisecond response time: The system uses efficient algorithm design to complete complex path planning calculations within milliseconds. Especially in the face of dynamically changing environments, the system can update and recalculate the path in real time to ensure that the vehicle always travels on the optimal route.

[0139] Parallel task scheduling: The system supports the scheduling and path optimization of multiple tasks in parallel. Through the distributed computing architecture, the scheduling system can handle multiple transportation tasks at the same time and provide independent optimal path planning for different vehicles. The efficiency of scheduling does not decrease significantly with the increase in the number of tasks.

[0140] 4.2 Real-time adjustment and task allocation

[0141] The system can not only quickly complete route planning, but also automatically adjust task allocation according to real-time changing environment. The system monitors the status of all transport vehicles and makes dynamic adjustments during the task according to the priority, urgency and road conditions of the task. For example, in a high-priority task, if the vehicle is delayed due to changes in the external environment (such as traffic jams or worsening weather), the system will re-plan tasks for other backup vehicles in real time to ensure that the overall transportation efficiency is not affected.

[0142] Automatic task reallocation: The system automatically allocates new tasks based on vehicle status and task urgency, and re-dispatches spare vehicles to ensure task progress.

[0143] Priority mechanism: The system has a built-in task priority evaluation model to prioritize efficient routes and transportation resources for urgent transportation tasks to ensure maximum optimization of transportation time.

[0144] 4.3. Improve security

[0145] 4.3.1 Multi-dimensional data monitoring

[0146] The system integrates multi-dimensional data such as vehicles, goods, external environment, space, and time through the Internet of Things technology to conduct all-round real-time monitoring of the transportation process. Each transport vehicle is equipped with a variety of sensors, and the system obtains data from these sensors in real time, including the vehicle's speed, location, and cargo status (such as temperature and pressure). In addition, external data sources (such as weather forecasts, traffic conditions), transportation space, and transportation time are also integrated into the system to provide more comprehensive data support for risk assessment.

[0147] Real-time vehicle monitoring: The status of each vehicle (such as running speed, braking condition, fuel usage, etc.) is uploaded to the system platform in real time. The system can monitor any abnormal conditions of the vehicle at the millisecond level to ensure early warning of potential problems.

[0148] Cargo status monitoring: The system will continuously monitor the physical status of hazardous chemicals (such as temperature, pressure, volatile gases, etc.) and automatically trigger emergency response measures when any abnormal changes are detected.

[0149] 4.3.2 Risk grading mechanism

[0150] The system has designed a risk grading mechanism to assess the risk of each transport task in real time. This mechanism not only takes into account the current transport environment (such as roads, weather, and regions), but also integrates vehicle status, cargo characteristics, and changes in the external environment. When the system detects an increase in the risk score, it immediately activates the emergency plan and replans the route.

[0151] Dynamic risk assessment: The risk grading mechanism is updated in real time, using vehicle sensors and external data sources to assess the potential risks of a mission. When the score rises to the medium-to-high risk range, the system will trigger an emergency plan in advance to reduce the possibility of an accident.

[0152] Automated emergency response: When an emergency occurs (such as vehicle failure or cargo abnormality), the system responds immediately and activates the corresponding emergency mechanism, such as re-planning the route and notifying the emergency department, to ensure maximum transportation safety.

[0153] 4.4 Data-driven optimization

[0154] 4.4.1. Historical data analysis

[0155] The system will store all the data of each transport task (such as vehicle operation status, cargo status, external environment, task execution time, etc.) and conduct detailed analysis after the task is completed. Through big data analysis technology, the system can identify potential optimization points in scheduling, such as which routes are usually congested, which types of cargo are more likely to have problems during transportation, etc. These historical data provide rich data support for future scheduling optimization.

[0156] Path history analysis: The system analyzes path selection in past transportation tasks, identifies which paths have higher risks (such as traffic congestion and frequent accidents), and uses this historical information for future path planning.

[0157] Task history optimization: Based on historical data such as task completion time and task success rate, the system can automatically optimize the scheduling model, predict possible delays in advance, and make preventive adjustments to future tasks.

[0158] 4.4.2 AI and machine learning optimization models

[0159] The system has built-in AI and machine learning modules, which can continuously learn and optimize based on historical data. After each scheduling task is completed, the system will analyze the completion of the task and automatically optimize the scheduling parameters through machine learning algorithms, such as adjusting the priority weights of different tasks and optimizing the weight settings in path planning.

[0160] Predictive scheduling: The system predicts possible bottlenecks in future tasks based on historical data, such as certain sections of road being frequently congested during certain periods of time or certain types of goods being more prone to abnormalities during transportation. Through predictive scheduling, the system can avoid potential problems in advance and improve the success rate of tasks.

[0161] Algorithm self-optimization: The AI ​​algorithm continuously adjusts the parameters of path planning and scheduling decisions based on the success rate and efficiency of historical tasks, ensuring that scheduling efficiency improves as data accumulates.

[0162] 4.5. Reduce management costs

[0163] 4.5.1. Automated dispatching system

[0164] By introducing an intelligent dispatching system, many decision-making processes in transportation tasks are automated, reducing the need for human intervention. The system can automatically complete a series of operations such as route planning, task allocation, and emergency response based on real-time data, greatly reducing the workload of dispatchers and reducing potential losses caused by human errors.

[0165] Reduce manual intervention: The system automatically handles processes such as route planning, task adjustment and emergency response. Dispatchers only need to intervene in special circumstances, greatly reducing the burden of manual operations.

[0166] 4.5.2. Cost optimization of emergency response

[0167] Through the intelligent emergency response system, when an emergency occurs, the system can respond quickly and take corresponding measures (such as re-planning routes, deploying spare vehicles, etc.), avoiding the lag of manual emergency response, and through real-time monitoring and early warning mechanisms, avoiding many risks that may lead to accidents in advance.

[0168] Faster emergency response: Emergency mechanisms are automatically triggered and executed in real time, significantly reducing delays in manual emergency response and potential losses.

[0169] Reduce unexpected losses: Since the system can quickly identify and respond to emergencies, it reduces losses caused by unexpected events (such as accidents, cargo damage, etc.), thereby indirectly reducing management costs.

[0170] 4.5.3 Systematic operation management

[0171] The system realizes the systematization and efficiency of transportation operations through centralized management of dispatching, route planning, emergency handling and other tasks. The intelligent dispatching platform not only improves operational efficiency through automation and intelligent means, but also reduces the waste of resources caused by manual operations and optimizes the resource allocation of the entire transportation system.

[0172] Efficient resource utilization: The system can automatically allocate vehicles and resources according to mission requirements, reducing the utilization rate of idle vehicles and operating costs.

[0173] Resource allocation optimization: Through intelligent monitoring of vehicle and cargo status, the system can dynamically adjust resource allocation, improve resource utilization efficiency, and avoid unnecessary waste.

[0174] Through the above in-depth refinement of dispatch efficiency, safety, data optimization and management costs, the intelligent dispatch system can significantly improve the overall performance of hazardous chemicals transportation and ensure efficient, safe and low-cost operation of the mission.

[0175] Implementation method 1: Intelligent scheduling platform architecture

[0176] The intelligent dispatching platform is the core of this system and is responsible for the management of the entire dispatching process, including vehicle status monitoring, risk assessment, route optimization and emergency response. The following is a detailed description of the functions of each module:

[0177] Vehicle monitoring module

[0178] Function description: The vehicle monitoring module is the basic module in the system, responsible for collecting dynamic information of transport vehicles in real time, monitoring the operating status of each vehicle, and processing data collected by sensors.

[0179] Data collection type:

[0180] Location data: Real-time monitoring of the vehicle's current location, driving speed and driving route through GPS.

[0181] Vehicle status data: The vehicle's operating status is collected through on-board sensors, including fuel consumption, engine temperature, brake system health, load capacity, etc.

[0182] Cargo monitoring: Real-time acquisition of cargo status (such as temperature, pressure, etc.) in hazardous chemicals transport vehicles to ensure cargo safety during transportation.

[0183] Data transmission: All data is transmitted to the system's central database via the Internet of Things for use in subsequent risk assessment and route optimization modules.

[0184] Abnormal monitoring: The system will continuously evaluate the vehicle status, detect abnormal conditions (such as excessive speed, abnormal braking, etc.), and issue an alarm in time.

[0185] Risk Assessment Module

[0186] Functional description: The risk assessment module calculates the current risk level of the transportation task based on real-time collected data and combined with the risk assessment model.

[0187] Risk Assessment Model:

[0188] Based on the health status of the vehicle, the external environment (such as roads, weather), the dangerousness of the cargo and time limits, the system automatically assigns weights for evaluation and generates a comprehensive risk score R.

[0189] The risk score of each task will be updated in real time. The risk level is divided into three levels: low, medium and high. The risk value is recalculated every few seconds.

[0190] Task classification: Based on the risk score, the system automatically divides tasks into three levels:

[0191] Low risk: The task is carried out as planned.

[0192] Medium risk: The system increases monitoring frequency and prepares backup plans.

[0193] High risk: The system immediately activates the emergency plan and replans the route to ensure mission safety.

[0194] Path optimization module

[0195] Functional description: The path optimization module calculates and adjusts the optimal driving route based on real-time traffic data, environmental conditions and task requirements through the dynamic A* path planning algorithm.

[0196] Dynamic path planning:

[0197] Real-time traffic monitoring: Obtain current traffic data (such as congestion, accidents) and adjust path weights in real time based on this information.

[0198] Environmental condition assessment: The system continuously monitors meteorological data (such as heavy rain, snow, wind speed, etc.), assigns different priorities to different road sections, and weights road risks.

[0199] Time optimization: Considering the urgency of the task, the route is optimized to ensure that the task is completed within the specified time. For time-sensitive tasks, the system gives priority to faster routes.

[0200] Real-time adjustment mechanism:

[0201] Every second, the system will recheck whether the current path is blocked (such as a new accident or weather change). If a sudden change in the environment is detected, the system will immediately re-plan a safe driving route and notify the driver in real time.

[0202] Emergency Response Module

[0203] Functional description: The emergency response module is automatically activated when a high-risk situation is detected to ensure the safety of hazardous chemicals during transportation.

[0204] Automatic warning system: When the vehicle sensor detects an abnormal state (such as brake system failure, abnormal cargo status, etc.), or the risk score reaches a high risk level, the system will automatically trigger an alarm and notify relevant personnel.

[0205] Implementation of emergency plan:

[0206] Route adjustment: Based on the current situation, the system will re-plan the safest route and inform the driver what to do.

[0207] Emergency department linkage: The system will notify relevant emergency departments (such as fire and public security) to prepare for potential accidents.

[0208] Safe parking solution: If the vehicle needs to stop in an emergency, the system will guide the driver to the nearest safe parking spot and notify the emergency team.

[0209] Implementation method 2: real-time scheduling application on mobile terminal

[0210] The mobile application is an important part of the system, allowing dispatchers and drivers to view task status and route information in real time, and provide quick feedback and adjustments when emergencies occur.

[0211] Driver-side functions

[0212] Real-time route display: Drivers can view the system-planned driving route in the app. The route display will be dynamically updated based on real-time traffic, weather and other factors to ensure that the driving route is always optimal.

[0213] Task status feedback: The driver can provide feedback on the current status of the vehicle (such as arrival time, current speed, etc.) through the application, and the application will also display the task completion progress based on the priority of the current task.

[0214] Risk Tips and Alerts:

[0215] When the system detects potential risks, the application will push an alert prompt through the notification bar, such as "An accident has occurred on the road ahead, it is recommended to adjust the route."

[0216] When the driver detects an abnormal situation, he can immediately send feedback through the application, and the system will re-evaluate the task and adjust the scheduling plan based on the feedback.

[0217] Emergency response function: The driver can trigger the emergency button through the application, upload the vehicle status and current location to the emergency response center, and the system will activate the emergency plan.

[0218] Dispatcher-side functions

[0219] Mission status monitoring: Dispatchers can view the mission status of all vehicles in real time through the application, including driving routes, mission priorities, vehicle status and real-time location. All data is displayed through a graphical interface, allowing dispatchers to understand the transportation progress at a glance.

[0220] Dynamic path adjustment:

[0221] The system automatically adjusts the vehicle's route when it detects changes in road conditions or the environment. Dispatchers can view route changes in real time and communicate with drivers to ensure that they follow up on the latest route adjustments in a timely manner.

[0222] Emergency Response Management:

[0223] When a mission reaches a high risk level, the dispatcher can trigger a manual emergency response through the app to notify the emergency department for support. The dispatcher can also directly adjust the route as needed or assign the mission to other backup vehicles.

[0224] By interacting with drivers in real time, dispatchers can monitor every stage of the mission and send new instructions through the app when necessary.

[0225] Mobile push and reminder functions

[0226] Smart push notifications: When the mission status or environment changes, the system will automatically send notifications based on the set thresholds to ensure that dispatchers and drivers are informed of the latest information in a timely manner. For example, "The road ahead is congested, please adjust your driving speed" or "The mission status has been updated, and the estimated arrival time has been delayed by 10 minutes."

[0227] Automatic reminders: The system will automatically generate reminders based on the urgency of the task to help the driver complete the task on time. If the task is delayed, the system will remind the driver to complete the important task first and provide alternative route suggestions.

[0228] Data visualization interface

[0229] Graphical display of routes: The route planning in the application is displayed in the form of a map, and drivers and dispatchers can view route changes at any time. The map displays real-time traffic conditions, road events and other information to help drivers understand the current environment.

[0230] Task status chart: The system will display the task completion progress in the form of a chart, including completion percentage, estimated arrival time and risk level of the current task, so that dispatchers can make quick judgments.

[0231] Through the combination of intelligent scheduling platform and mobile applications, the system can monitor hazardous chemical transportation tasks in real time to ensure efficient, flexible and safe route planning and scheduling.

Claims

1. A real-time dynamic dispatching system for hazardous chemicals vehicles in chemical parks based on multi-dimensional data fusion and intelligent decision-making, characterized in that: include: The vehicle monitoring module is used to collect dynamic information of transport vehicles in real time, monitor the operating status of each vehicle, and process the data collected by sensors; The risk assessment module is used to calculate the current risk level of the transportation task based on the real-time collected data combined with the risk assessment model; Path optimization module, which is used to calculate and adjust the optimal driving route based on real-time traffic data, environmental conditions and mission requirements through the dynamic A* path planning algorithm; The emergency response module is used to automatically start when a high-risk situation is detected to ensure the safety of hazardous chemicals during transportation.

2. The real-time dynamic dispatching system for hazardous chemicals vehicles in chemical parks based on multi-dimensional data fusion and intelligent decision-making according to claim 1 is characterized in that: The data collection types of the vehicle monitoring module include: Location data: Real-time monitoring of the vehicle’s current location, driving speed and driving route through GPS; Vehicle status data: collect the vehicle's operating status through on-board sensors; Cargo monitoring: Real-time acquisition of cargo status in hazardous chemicals transport vehicles to ensure cargo safety during transportation; Data transmission: All data is transmitted to the system central database through the Internet of Things for use by subsequent risk assessment and route optimization modules; Abnormal monitoring: The system will continuously evaluate the vehicle status, detect abnormal conditions, and issue an alarm in time.

3. The real-time dynamic dispatching system for hazardous chemicals vehicles in chemical parks based on multi-dimensional data fusion and intelligent decision-making according to claim 1 is characterized in that: The risk assessment module includes building a risk assessment model: Based on the vehicle's health status, external environment, cargo danger and time limit, the system automatically assigns weights for evaluation and generates a comprehensive risk score R; The risk score of each task is updated in real time. The risk level is divided into three levels: low, medium, and high. The risk value is recalculated every few seconds. Task classification: Based on the risk score, the system automatically divides tasks into three levels: Low risk: The task is carried out as planned; Medium risk: The system increases monitoring frequency and prepares backup plans; High risk: The system immediately activates the emergency plan and replans the route to ensure mission safety.

4. The real-time dynamic dispatching system for hazardous chemicals vehicles in chemical parks based on multi-dimensional data fusion and intelligent decision-making according to claim 3 is characterized in that: The risk assessment model has the following assessment methods: The following multidimensional risk factors were established: Cargo Hazard Rating H: Rating is based on the category, characteristics, storage temperature and pressure of hazardous chemicals; including but not limited to flammable, explosive and toxic substances; Vehicle Status Score V: A comprehensive assessment of the vehicle's current operating status, including health status, maintenance records, real-time monitoring data, especially vehicle failure rate, driver experience, and vehicle load; Environmental risk score E: Based on the assessment of external environmental factors such as real-time weather conditions, traffic flow, road conditions, and air quality. Bad weather, congested roads, and polluted air quality will increase this score. Spatial risk score S: The vehicle is scored based on whether its route passes through high-risk areas. The risk score is higher if it passes through these areas. Time risk score T: It is evaluated based on the time urgency of the transportation task and whether it is carried out during peak hours or special times. The score increases when the time is urgent or the transportation is carried out during special periods; A nonlinear weighting function is adopted, and a cross factor is introduced to reflect the mutual influence between different risk factors; in addition, since the specific situation of the task will change dynamically, the weight of each factor, α, β, γ, δ,∈, is dynamically adjusted and can be updated at any time; The risk score formula is: 1) Normalization factor Normalization is to unify each factor to the range of 0-1 to solve the problem of different magnitudes between factors; 2) Linear terms and logarithmic terms The linear term on the left side of the plus sign: represents the direct impact of each factor on the risk score, which is used to reflect the importance of each factor; The logarithmic term on the right side of the plus sign is used to deal with the risk amplification effect in extreme cases. The growth rate of the logarithmic term is fast at first and then slows down. It can significantly improve the risk score when the factor value is small, but when the factor value is high, the amplification effect gradually weakens. 3) Index weighting of spatial risk The spatial risk score S reflects the risk of a vehicle passing through a specific area; the exponential form is used to describe the amplification effect of this risk on the overall score; 4) Linear and nonlinear terms of time risk ∈1×T 归一化 +∈2×T 2 The linear term on the left side of the plus sign: describes the direct impact of time risk on the overall score; Nonlinear term on the right side of the plus sign: When the time risk score T exceeds the normal range, the square term will quickly amplify the impact of the time factor on the overall score; 5) Cross-term η1×T×E+η2×H×S The interaction term between time and environment on the left side of the plus sign: reflects the joint effect of time pressure and environmental risk; The cargo and space interaction term on the right side of the plus sign reflects the amplification effect of dangerous goods transportation passing through high-risk areas; 6) Risk level classification Based on the improved risk score R, the system will evaluate each transportation task in real time, determine its risk level, and trigger the corresponding scheduling strategy; the evaluation results are divided into three risk levels: Low risk R<30: The task is performed according to the normal route and no additional intervention is required; Medium risk 30≤R<60: At this time, the task needs to increase the monitoring frequency, and the system will dynamically adjust the path to avoid potential high-risk areas; at the same time, prepare backup plans to deal with possible emergencies. High risk R ≥ 60: The system triggers the emergency plan, replans the route, prioritizes avoiding high-risk areas, and notifies relevant emergency departments to ensure the safe completion of the mission.

5. The real-time dynamic dispatching system for hazardous chemicals vehicles in chemical parks based on multi-dimensional data fusion and intelligent decision-making according to claim 1 is characterized in that: The path optimization module includes: Dynamic path planning: Real-time traffic monitoring: obtain current traffic data and adjust path weights in real time based on this information; Environmental condition assessment: The system continuously monitors meteorological data, assigns different priorities to different road sections, and weights road risks; Time optimization: Considering the urgency of the task, the system optimizes the route to ensure that the task is completed within the specified time. For time-sensitive tasks, the system gives priority to faster routes. Real-time adjustment mechanism: Every second, the system will recheck whether the current path is blocked. If a sudden change in the environment is detected, the system will immediately re-plan a safe driving route and notify the driver in real time.

6. The real-time dynamic dispatching system for hazardous chemicals vehicles in chemical parks based on multi-dimensional data fusion and intelligent decision-making according to claim 5 is characterized in that: The dynamic A* path planning algorithm is used, and scheduling adjustments are made in combination with real-time data to ensure that the vehicle travels on the optimal path; including the path optimization algorithm, the specific steps are as follows: Initialize path planning: The system initializes the A* path planning algorithm based on the vehicle's departure and destination, as well as the current real-time environmental data, and calculates the initial optimal path; the execution process is: First, the vehicle’s GPS is used to obtain the current location and the designated destination coordinates, and the GIS system is used to obtain map data of all possible routes; Initialize the A* algorithm, set the starting node n_start and the target node n_goal, divide the path search space into multiple small nodes, and perform a preliminary search on all path nodes; Starting from the initial node, the calculation is started according to the principle of shortest distance, and potential obstacles and traffic conditions on the path are evaluated; Weight setting: weights are assigned based on road conditions, traffic flow, external environment, space and time factors. Dynamic weights are assigned to each path node based on different factors. Road sections with lower weights are given priority. That is, an initial weight is set for each path node, and the weight is dynamically adjusted based on factors such as road conditions, traffic flow, external environment, regional risks, and time requirements; Heuristic function: The algorithm uses the heuristic function h to estimate the cost from node n to the target node, taking into account the length of the road section, traffic conditions, and risk areas. The cost of the path nodes is evaluated through the heuristic function to ensure that the optimal path is selected; the execution process is: In the A* algorithm, the heuristic function h is used to estimate the cost from node n to the target node; The calculation method combines the path length, road conditions and risk assessment results; h(n)=d(n,n_goal+∑risk(n) (2) Where d(n,n_goal) is the distance from node n to target node n_goal, ∑risk(n) is the cumulative risk assessment value of the current path node; risk assessment not only considers the length of the road section, but also combines comprehensive factors including spatial risk and time risk; Path search and evaluation: By expanding the preferred path nodes, selecting the optimal path and updating the path weight according to real-time data, the vehicle is ensured to travel on the safest route; the execution process is as follows: Starting from the starting node n_start, the path is gradually expanded along the node with the smallest weight; each time the algorithm expands, it will select the next node with the lowest path cost and give priority to exploring a better path; Monitor the weight update of all nodes in real time. If the risk assessment score of a path exceeds the set threshold, the A* algorithm will exclude this path from the optimal path candidate. Continuously evaluate the security, time cost, and external environmental impact of the expansion path to ensure that the path remains the best choice during the expansion process; Dynamic adjustment: When the system detects changes in the external environment, the A* algorithm will automatically replan the path to avoid the affected area; the execution process is: Update the data of all path nodes every second to evaluate whether there are sudden changes; Once an abnormality in the environment or vehicle status is detected, the A* algorithm will recalculate and update the path in real time to ensure that the vehicle can adjust to a safe driving route in the shortest time.

7. The real-time dynamic dispatching system for hazardous chemicals vehicles in chemical parks based on multi-dimensional data fusion and intelligent decision-making according to claim 6 is characterized in that: It also includes a scheduling adjustment mechanism, the specific steps are as follows: Real-time data monitoring: The system collects real-time data of vehicles, goods, external environment, space and time every second, and updates the evaluation model based on the data; the execution process is as follows: Collect real-time data from vehicle sensors every second and integrate it with external environmental data, spatial data, and temporal data through the IoT platform; Based on real-time data, the system continuously updates the vehicle’s current status and compares it with the expected data from the evaluation model; If the data is abnormal, the system will mark the task as a potential risk; Detect risk factors: Identify and evaluate risk factors that affect vehicle transportation. If the system finds that the risk score rises to medium or high risk, it will immediately enter the scheduling adjustment process; The execution process is: The system inputs multi-dimensional data into the assessment model and calculates the real-time risk score R; when the score exceeds the set threshold, the system will enter the scheduling adjustment mode; Through the calculation of the aforementioned risk assessment model, combined with the dangerousness of the cargo, the vehicle status, the external environment, the space restrictions and the time restrictions, multiple scores are comprehensively calculated to obtain the current risk level of the task; The threshold of the risk assessment score is adjusted dynamically to ensure that potential problems can be discovered in advance in special emergency situations; Path replanning: When the risk score increases, the system quickly adjusts the path planning to ensure safety. The system recalculates the optimal path through the dynamic A* algorithm and notifies the driver of the new route. The execution process is as follows: When the system detects that a mission is at medium or high risk, it immediately initiates dynamic A* path recalculation; based on the latest real-time data, the system regenerates a route that avoids high-risk areas or traffic congestion; After the new route is generated, the system pushes the updated route to the driver through the vehicle system or mobile application, and sends risk alerts to the dispatcher to ensure that the driver and dispatcher can respond quickly when the task status changes; Emergency response trigger: In high-risk conditions, the emergency plan is activated to ensure transportation safety. If the risk score reaches high risk, the system triggers the emergency plan, notifies the emergency department and guides the vehicle to a safe area; the execution process is: When the system detects that the risk score R reaches a high risk, the system will immediately trigger the emergency response mechanism; first, the system will notify the nearest emergency command center and relevant emergency departments to make safety preparations in advance; At the same time, the system will plan a route for the vehicle to reach the nearest safe zone and notify the driver of the latest route through the mobile terminal; The emergency response process also includes urgent monitoring of the transported goods. If necessary, the system will require the driver to stop midway to prevent further harm; In extreme cases, the system will also arrange a special emergency fleet or rescue force for rapid response.

8. The real-time dynamic dispatching system for hazardous chemicals vehicles in chemical parks based on multi-dimensional data fusion and intelligent decision-making according to claim 1 is characterized in that: The emergency response module includes: Automatic warning system: When the vehicle sensor detects an abnormal state or the risk score reaches a high risk level, the system will automatically trigger an alarm and notify relevant personnel; Path adjustment: Based on the current situation, the system will re-plan the safest route and inform the driver how to operate; Emergency department linkage: The system will notify relevant emergency departments to prepare for potential accidents; Safe parking solution: If the vehicle needs to stop in an emergency, the system will guide the driver to the nearest safe parking spot and notify the emergency team.

Images