Safety risk evaluation and analysis system based on van

Abstract

The application discloses a minibus-based safety risk evaluation and analysis system and relates to the technical field of safety risk evaluation.The application comprises a user risk analysis module, a driving risk analysis module and a pre-warning supervision analysis module.The application firstly constructs a vehicle body aging index model, a driving aggressiveness index model and a line risk index model through the user risk analysis module, secondly collects current risk-related data of the user through the driving risk analysis module, inputs the current risk-related data into the vehicle body aging index model, the driving aggressiveness index model and the line risk index model, obtains current indexes, finally determines the user risk grade and pre-warns through the pre-warning supervision analysis module, generates a risk heat map and a trend analysis report, and statistically analyzes the situation in a region to clearly show the risk region distribution and a high-risk driving behavior sequence, and finally formulates a regional supervision scheme, so that the safety risk of the minibus is comprehensively evaluated and effectively supervised.

Description

Technical Field

[0001] This invention relates to the field of safety risk assessment technology, specifically to a safety risk assessment and analysis system based on a minivan. Background Technology

[0002] Minivans are typically used for both passenger and cargo transport, and some vehicles are used frequently, making maintenance and upkeep easy to overlook, resulting in prominent issues such as vehicle body aging and component wear. At the same time, their driving scenarios are diverse, and they face complex road and environmental risks, requiring a targeted risk assessment mechanism. Therefore, a safety risk assessment and analysis system based on minivans is needed.

[0003] Traditional vehicle safety management relies heavily on manual inspections or general risk assessment systems, which do not fully consider the specific risks of minivans. This results in insufficient accuracy in risk assessment and difficulty in effectively identifying safety hazards unique to minivans. Furthermore, the safety risks of minivans involve multiple dimensions of factors, including the vehicle, people, roads, and the environment. Existing systems often struggle to integrate multi-source data for real-time analysis and cannot provide timely warnings of potential risks, leading to weak accident prevention capabilities.

[0004] Current technologies for traffic safety risk assessment mostly focus on assessments under single or limited factors, lacking comprehensive and accurate methods for assessing multiple risk factors. Some studies only consider vehicle operation and accident information, neglecting the impact of drivers on traffic; others rely on partial indicators as risk judgment criteria, ignoring the timely collection, assessment, and alerting of external environmental safety information.

[0005] Existing technologies are insufficient to provide regulatory authorities with comprehensive and targeted decision support based on risk assessment results. They lack modules for generating risk heat maps and trend analysis reports to analyze the distribution of risk areas and sequences of high-risk driving behaviors, thereby enabling the development of detailed regional regulatory plans. Furthermore, they lack measures such as setting up additional enforcement points and monitoring equipment in high-risk areas and conducting special rectification campaigns targeting high-risk driving behaviors. Consequently, they cannot effectively improve the effectiveness and accuracy of regional regulation and fail to meet the actual needs of regulatory authorities for the safety management of minibuses. Summary of the Invention

[0006] To address the aforementioned technical shortcomings, the purpose of this invention is to provide a safety risk assessment and analysis system based on minivans.

[0007] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides a safety risk assessment and analysis system based on a minivan, including the following modules: a user risk analysis module, used to collect the user's historical vehicle driving data, historical driving behavior data and historical external interface data, analyze the user's historical vehicle driving data to obtain a vehicle body aging index risk model, analyze the user's historical driving behavior data to obtain a driving aggression index risk model, and analyze the user's historical external interface data to obtain a route risk index model.

[0008] The driving risk analysis module is used to collect the user's current vehicle driving data, current driving behavior data, and current external interface data. It analyzes the user's current vehicle driving data, current driving behavior data, and current external interface data to obtain the user's current vehicle aging index, current driving aggression index, and current route risk index.

[0009] The early warning and monitoring analysis module is used to analyze the user's current vehicle aging index, current aggressive driving index, and current route risk index to obtain the user's risk level, regional risk heat map, and regional trend analysis report. It issues user warnings based on the user's risk level, and further analyzes the risk area distribution and high-risk driving behavior sequences based on the regional risk heat map and trend analysis report, ultimately generating a regional monitoring plan.

[0010] Preferably, the analysis obtains the user's risk level, the region's risk heat map, and the region's trend analysis report. The specific analysis process is as follows: multiply the user's basic risk index by the personal risk adjustment factor to obtain the user's risk index; retrieve the risk index range corresponding to each risk level from the database; if the user's risk index belongs to a certain risk index range, it indicates that the user's risk level is that risk level.

[0011] By using GPS positioning to obtain the risk level and location of each user in the area, the risk level of each vehicle is assigned a different color, and the data is marked and aggregated on an electronic map to form a heat map of risk distribution in the area.

[0012] Based on the user's risk adjustment factor, the risk adjustment factor of each user in the region is obtained. Users whose personal risk adjustment factor is greater than the preset risk adjustment factor are recorded as effective trend users. Based on the characteristic parameters of each violation behavior of each effective trend user in the region, the type of each violation behavior of each effective trend user in the region is obtained through feature recognition technology. The occurrence frequency of each type of violation in the region is counted, thereby obtaining the occurrence frequency of each type of violation behavior in each statistical time period in the region. A coordinate system is established to obtain the trend graph of the change of each type of violation behavior over time, thereby obtaining a trend analysis report.

[0013] The beneficial effects of this invention are as follows: 1. This invention first constructs a vehicle body aging index model, a driving aggression index model, and a route risk index model through a user risk analysis module. Secondly, it collects current risk-related data of users through a driving risk analysis module and inputs it into the vehicle body aging index model, the driving aggression index model, and the route risk index model to obtain the current indices. Finally, it determines the user's risk level and issues an early warning through an early warning and supervision analysis module, generates a risk heat map and a trend analysis report, and statistically analyzes the situation within the region to clarify the distribution of risk areas and the sequence of high-risk driving behaviors. Finally, it formulates a regional supervision plan to achieve a comprehensive evaluation and effective supervision of the safety risks of minivans.

[0014] 2. By collecting multi-source data in real time through the vehicle terminal and combining it with machine learning algorithms for rapid processing and analysis, the system can output risk levels in real time, trigger audible and visual alarms or SMS notifications, enabling businesses and drivers to promptly grasp vehicle and driving risks, avoid potential dangers in advance, and effectively reduce the incidence of traffic accidents.

[0015] 3. The risk heat map, trend analysis report, and risk area distribution data generated by the system can intuitively reflect the spatial distribution and temporal change patterns of van safety risks, assisting regulatory authorities in identifying high-risk areas and high-frequency risk behaviors, formulating targeted governance strategies, adding enforcement points, carrying out special rectification campaigns, and improving the scientific and precise nature of supervision. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art 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.

[0017] Figure 1 This is a schematic diagram of the system structure connection of the present invention. Detailed Implementation

[0018] 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.

[0019] according to Figure 1 As shown, the present invention provides a safety risk assessment and analysis system based on a minivan, including the following modules: a user risk analysis module, a driving risk analysis module, an early warning and monitoring analysis module, and a database.

[0020] The driving risk analysis module is connected to the user risk analysis module and the early warning and monitoring analysis module, respectively. The user risk analysis module, driving risk analysis module, and early warning and monitoring analysis module are all connected to the database.

[0021] The user risk analysis module is used to collect users' historical vehicle driving data, historical driving behavior data, and historical external interface data. It analyzes users' historical vehicle driving data to obtain a vehicle body aging index risk model, analyzes users' historical driving behavior data to obtain a driving aggression index risk model, and analyzes users' historical external interface data to obtain a route risk index model.

[0022] In one specific embodiment, the user's historical vehicle driving data, historical driving behavior data, and historical external interface data are collected. The specific collection process is as follows: The user's historical vehicle driving data includes the frequency of engine fault codes, cumulative mileage, maintenance interval, vehicle age, and accident characteristic parameters, which are collected in real time through the OBD module integrated in the vehicle terminal and connected to the vehicle's OBD interface. The OBD module can directly read the frequency of engine fault codes and cumulative mileage stored in the vehicle's ECU. The maintenance interval is obtained by manually entering or uploading vehicle maintenance records on the real-time monitoring platform. The vehicle's age is obtained by entering the vehicle's initial registration date. Accident characteristic parameters are collected through external data interfaces and the vehicle terminal.

[0023] It should be noted that accident characteristic parameters include, but are not limited to, the degree of damage caused by the accident, the vehicle speed at the moment of the accident, and the braking status at the moment of the accident.

[0024] The user's historical driving behavior data includes the number of sudden accelerations and their values, speeding duration, characteristic parameters of each violation, continuous driving time, and cumulative driving time. The acceleration sensor deployed in the vehicle terminal collects the number of sudden accelerations and their values. The GPS locator integrated in the vehicle terminal obtains the vehicle speed and location. Based on the vehicle location, the corresponding vehicle speed threshold is obtained to determine whether speeding has occurred, and the speeding duration is then calculated. At the same time, the continuous driving time and cumulative driving time are obtained through the GPS locator. The camera deployed in the vehicle terminal collects driver behavior data. By identifying the driver behavior data, violations are determined, and the behavioral data of the violations are recorded as characteristic parameters of the violations.

[0025] It should be noted that the characteristic parameters of violations include, but are not limited to, the driver's hand position, head turning angle, seat belt position, and seat belt tension. Violations include, but are not limited to, making or receiving phone calls and not wearing a seat belt.

[0026] The user's historical external interface data includes weather feature data, road congestion index, and traffic violation record feature data corresponding to the historical driving route. The route location is obtained based on the historical driving route recorded by the vehicle terminal GPS locator. The weather feature data is obtained by connecting to the historical meteorological database of the meteorological department. The average vehicle speed and speed limit of the road segment corresponding to the route location are obtained through the external traffic data platform interface. The congestion index is obtained by dividing the average vehicle speed of the road segment by the speed limit. The traffic violation record feature data is collected by connecting to the traffic management department's traffic violation record system interface.

[0027] It should be noted that weather characteristic data includes, but is not limited to, weather types and levels such as rainfall, fog, high temperature, and strong wind. Traffic violation record characteristic data includes, but is not limited to, the time difference between the red light duration and the vehicle crossing the line, the duration of continuous occupation of the emergency lane, and the duration of speeding.

[0028] In one specific embodiment, the analysis of the user's historical vehicle driving data is carried out as follows: based on the cumulative mileage and frequency of engine fault codes in each historical mileage period, the frequency of fault codes in each mileage interval is statistically obtained. A coordinate axis is established with mileage as the horizontal axis and the frequency of fault codes as the vertical axis. After fitting the curve, the curve of fault frequency as mileage increases is obtained, and the slope of fault frequency in each mileage interval is obtained.

[0029] Based on the vehicle's age, the standard maintenance interval for each vehicle's age is retrieved from the database. This standard maintenance interval is then used to obtain the user's historical standard maintenance interval. The difference between the standard maintenance interval and the original maintenance interval is subtracted and divided by the original standard maintenance interval to obtain the maintenance interval deviation. This deviation is then used to obtain the user's historical maintenance interval deviation. Finally, based on the cumulative mileage from each historical mileage interval, the maintenance interval deviation for each mileage range is obtained.

[0030] A multiple regression model is adopted, with the slope of the fault frequency and the maintenance interval deviation in each mileage range as independent variables, and the accident characteristic parameters of each historical statistics in each mileage range as dependent variables for training. The model outputs the vehicle body aging index, thereby obtaining the vehicle body aging index risk model. This establishes a vehicle body aging index risk model that takes the slope of the fault frequency and the maintenance interval deviation as input and outputs the vehicle body aging index.

[0031] It should be noted that the multiple regression model is an existing technology and can be found on the Internet, so it will not be elaborated further.

[0032] In one specific embodiment, the analysis of the user's historical driving behavior data is carried out as follows: the user's historical rapid acceleration values ​​are divided by a preset rapid acceleration value threshold to obtain the risk value of each historical rapid acceleration; the sum of the risk values ​​of each historical rapid acceleration is divided by the number of rapid accelerations to obtain the user's historical rapid acceleration risk rate.

[0033] It should be noted that the rapid acceleration threshold is the acceleration threshold for normal speed changes. When the acceleration value is greater than the threshold, it indicates that the current driving is a sudden braking or rapid acceleration, which poses a risk of collision. The specific value is set by the staff.

[0034] Divide the overspeed duration by the cumulative driving time to obtain the overspeed rate, and use this to obtain the overspeed rate of each user's historical statistics.

[0035] The similarity analysis is performed between the feature parameters of each violation and the feature parameters of various violations in the database to obtain the similarity of each violation to various types of violations. The violation type with the highest similarity is selected as the type of each violation, thereby obtaining the violation type of each violation in the user's historical statistics. The violation risk index of each violation type is obtained from the database, and then the violation risk index of each violation is obtained. The total violation risk index of each user's historical statistics is calculated and divided by the total violation risk index threshold to obtain the violation risk rate of each user's historical statistics.

[0036] It should be noted that the total violation risk index threshold is the maximum total violation risk index during normal driving. When the total violation risk index exceeds the threshold, it indicates that there is a risk of violation during driving.

[0037] A weighted scoring model is established based on the user's historical statistics of acceleration risk rate, speed exceeding rate, and violation risk rate. This weighted scoring model is denoted as the driving aggression index risk model. A driving aggression index risk model is established that takes acceleration risk rate, speed exceeding rate, and violation rate as input and outputs driving aggression index.

[0038] It should be noted that the weighted scoring model is existing technology and can be found on the Internet, so it will not be elaborated further.

[0039] In one specific embodiment, the analysis of the user's historical external interface data is carried out as follows: data of different dimensions are converted into standardized values ​​to obtain weather feature risk values, road congestion risk values, and route traffic accident risk values. A random forest algorithm is used, with the route traffic accident risk values ​​occurring on historical routes as labels, to train the standardized weather feature risk values, road congestion risk values, and road risk values, and output a route risk index. This establishes a route risk index model that takes weather feature risk values ​​and road congestion risk values ​​as input and outputs a route risk index.

[0040] It should be noted that the random forest algorithm is an existing technology and can be found on the Internet, so it will not be described in detail here.

[0041] The driving risk analysis module is used to collect the user's current vehicle driving data, current driving behavior data, and current external interface data. It analyzes the user's current vehicle driving data, current driving behavior data, and current external interface data to obtain the user's current vehicle aging index, current driving aggression index, and current route risk index.

[0042] In one specific embodiment, the process of collecting the user's current vehicle driving data, current driving behavior data, and current external interface data is as follows: Based on the process of collecting the user's historical vehicle driving data, historical driving behavior data, and historical external interface data, the user's current vehicle driving data, current driving behavior data, and current external interface data are collected.

[0043] In one specific embodiment, the process of obtaining the user's current vehicle aging index, current driving aggressiveness index, and current route risk index is as follows: the user's current vehicle driving data, current driving behavior data, and current external interface data are respectively input into the driving aggressiveness index risk model, the driving aggressiveness index risk model, and the route risk index to obtain the user's current vehicle aging index, current driving aggressiveness index, and current route risk index.

[0044] The vehicle driving data, driving behavior data, and external interface data collected from each statistical session in the database are input into the driving aggression index risk model, driving aggression index risk model, and route risk index, respectively, to obtain the vehicle body aging index, driving aggression index, and route risk index collected by the user for each statistical session.

[0045] The threshold values ​​for vehicle aging index, driving aggression index, and route risk index are obtained from the user's statistical data on vehicle aging index, driving aggression index, and route risk index. Based on the user's current vehicle aging index, driving aggression index, and route risk index, each index is divided by a standard value and then weighted to obtain the user's basic risk index. The user's current vehicle aging index, driving aggression index, and route risk index are then divided by the threshold values ​​and weighted to obtain the user's personal risk correction index. The correction factor corresponding to each user's risk correction index is obtained from the database to obtain the user's personal risk correction factor.

[0046] It should be noted that the threshold values ​​for vehicle aging index, driving aggression index, and route risk index are the threshold values ​​for normal driving conditions determined by staff through historical data analysis. When the vehicle aging index, driving aggression index, or route risk index exceeds the threshold, the risk of an accident is higher. The specific values ​​are set by the staff. The weighted factors for the calculation are the vehicle aging weight factor, driving aggression weight factor, and route risk weight factor, which are obtained by staff through historical data analysis. The more frequent the van's malfunctions, the higher the vehicle aging weight factor; the more driver violations, the higher the driving aggression weight factor; and the more frequent the road accidents, the higher the route risk weight factor. The specific settings are determined by the staff.

[0047] The early warning and monitoring analysis module is used to analyze the user's risk level, regional risk heat map, and regional trend analysis report based on the user's current vehicle aging index, current aggressive driving index, and current route risk index. It issues user warnings based on the user's risk level, and statistically generates risk heat maps and trend analysis reports for each user in the region. Furthermore, it analyzes the distribution of risk areas and high-risk driving behavior sequences, and finally generates a regional monitoring plan.

[0048] In one specific embodiment, the analysis yields the user's risk level, the region's risk heatmap, and the region's trend analysis report. The specific analysis process is as follows: the user's basic risk index is multiplied by a personal risk adjustment factor to obtain the user's risk index. The risk index range corresponding to each risk level is obtained from the database. If the user's risk index belongs to a certain risk index range, it indicates that the user's risk level is that risk level.

[0049] By using GPS positioning to obtain the risk level and location of each user in the area, the risk level of each vehicle is assigned a different color, and the data is marked and aggregated on an electronic map to form a heat map of risk distribution in the area.

[0050] Based on the user's risk adjustment factor, the risk adjustment factor of each user in the region is obtained. Users whose personal risk adjustment factor is greater than the preset risk adjustment factor are recorded as effective trend users. Based on the characteristic parameters of each violation behavior of each effective trend user in the region, the type of each violation behavior of each effective trend user in the region is obtained through feature recognition technology. The occurrence frequency of each type of violation in the region is counted, thereby obtaining the occurrence frequency of each type of violation behavior in each statistical time period in the region. A coordinate system is established to obtain the trend graph of the change of each type of violation behavior over time, thereby obtaining a trend analysis report.

[0051] In one specific embodiment, the warning process is as follows: when the user's risk level is greater than the basic risk level, the user is notified via SMS; when the user's risk level is greater than the standard risk level, the user is notified via SMS and an audible and visual alarm is triggered.

[0052] In one specific embodiment, the analysis yields the risk area distribution and high-risk driving behavior sequences. The specific analysis process is as follows: the region is divided into grid units according to a preset scale, and each grid serves as the basic unit of spatial analysis. The risk level information of all vehicles within the grid is recorded, and the number and proportion of vehicles of each risk level within each grid are counted. A spatial clustering algorithm is used to aggregate grids with a density exceeding a threshold for each risk level to form continuous risk areas of each risk level. This yields the risk area boundaries for each risk level and obtains the risk area distribution.

[0053] The risk level density threshold is the maximum value of the normal risk level density. When the risk level density is greater than the threshold, it indicates that the corresponding risk level is the risk level of the area. The specific value is set by the staff, and is generally set to 0.5.

[0054] By using image recognition technology, the slope of various violations is obtained from the trend graph of various violations over time. The violations are then sorted in descending order of slope to obtain a sequence of high-risk driving behaviors.

[0055] In one specific embodiment, the generation process of the regional supervision scheme is as follows: For the distribution of risk areas: temporary law enforcement points and fixed monitoring equipment are added in risk areas where the risk level is higher than the standard risk level, and the priority value of each risk area in the patrol route is increased according to the risk level of each risk area.

[0056] For high-risk driving behavior sequences: a special rectification campaign will be launched, and each type of violation at the top of the sequence will be recorded as a high-risk driving behavior. Accident cases of high-risk driving behaviors will be sent to enterprises in the region to warn them to strengthen driver training. In addition, regional violation warning announcements will be issued in conjunction with traffic management departments to increase the monitoring density of corresponding violation types.

[0057] The database stores the standard maintenance intervals for each vehicle's service life, standard characteristic parameters for various violations, vehicle driving data collected by the user, driving behavior data collected by the user, external interface data collected by the user, and risk index ranges corresponding to each risk level.

[0058] The above description is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they should all fall within the protection scope of the present invention.

Claims

1. A safety risk assessment and analysis system based on minivans, characterized in that, Includes the following modules: The user risk analysis module is used to collect users' historical vehicle driving data, historical driving behavior data, and historical external interface data. It analyzes users' historical vehicle driving data to obtain a vehicle body aging index risk model, analyzes users' historical driving behavior data to obtain a driving aggression index risk model, and analyzes users' historical external interface data to obtain a route risk index model. The driving risk analysis module is used to collect the user's current vehicle driving data, current driving behavior data, and current external interface data. It analyzes the user's current vehicle driving data, current driving behavior data, and current external interface data to obtain the user's current vehicle aging index, current driving aggression index, and current route risk index. The early warning and monitoring analysis module is used to analyze the user's current vehicle aging index, current aggressive driving index, and current route risk index to obtain the user's risk level, regional risk heat map, and regional trend analysis report. It issues user warnings based on the user's risk level, and further analyzes the risk area distribution and high-risk driving behavior sequences based on the regional risk heat map and trend analysis report, ultimately generating a regional monitoring plan. The database stores the standard maintenance intervals for each vehicle's service life, standard characteristic parameters for various violations, vehicle driving data collected by the user, driving behavior data collected by the user, external interface data collected by the user, and risk index ranges corresponding to each risk level.

2. The safety risk assessment and analysis system based on a minivan according to claim 1, characterized in that, The analysis of the user's historical vehicle driving data is performed in the following specific process: The user's historical vehicle driving data includes the frequency of engine fault codes, cumulative mileage, maintenance interval, vehicle age and accident characteristic parameters, as statistically analyzed by the user in each historical period. Based on the cumulative mileage and frequency of engine fault codes in each user's historical statistics, the frequency of fault codes in each mileage interval is statistically obtained. A coordinate axis is established with mileage as the horizontal axis and the frequency of fault codes as the vertical axis. After fitting the curve, the curve of fault frequency as mileage increases is obtained, and the slope of fault frequency in each mileage interval is obtained, which is denoted as the slope of fault frequency in each mileage interval. Based on the vehicle's age, the standard maintenance interval for each vehicle's age is retrieved from the database. This standard maintenance interval is then obtained, and the user's historical standard maintenance interval is calculated. The difference between the standard maintenance interval and the original maintenance interval is subtracted from the original maintenance interval, and then divided by the original maintenance interval to obtain the maintenance interval deviation. This deviation is then calculated, and the user's historical maintenance interval deviation is calculated. Based on the cumulative mileage in each historical mileage calculation, the maintenance interval deviation for each mileage range is obtained. A multiple regression model is adopted, with the slope of the fault frequency and the maintenance interval deviation in each mileage range as independent variables, and the accident characteristic parameters of each historical statistics in each mileage range as dependent variables for training. The model outputs the vehicle body aging index, thereby obtaining the vehicle body aging index risk model. This establishes a vehicle body aging index risk model that takes the slope of the fault frequency and the maintenance interval deviation as input and outputs the vehicle body aging index.

3. The safety risk assessment and analysis system based on a minivan according to claim 2, characterized in that, The analysis of the user's historical driving behavior data is carried out in the following specific process: The user's historical driving behavior data includes the number of times the user experienced sudden acceleration and the value of each sudden acceleration, the duration of speeding, the characteristic parameters of each violation, the duration of continuous driving and the cumulative driving time. The number of occurrences of rapid acceleration and the value of each rapid acceleration in the user's historical statistics are normalized to obtain the rapid acceleration risk rate of the user's historical statistics. Divide the speeding duration by the cumulative driving time to get the speeding rate, and use this to obtain the speeding rate of each user's historical statistics. By performing similarity analysis between the characteristic parameters of each violation and the standard characteristic parameters of various violations in the database, the violation risk index of each violation is obtained. The total violation risk index of each user's historical violations is calculated and divided by the total violation risk index threshold to obtain the violation risk rate of each user's historical violations. A weighted scoring model is established based on the user's historical statistics of acceleration risk rate, speed exceeding rate, and violation risk rate. This weighted scoring model is denoted as the driving aggression index risk model. A driving aggression index risk model is established that takes acceleration risk rate, speed exceeding rate, and violation rate as input and outputs driving aggression index.

4. The safety risk assessment and analysis system based on a minivan according to claim 1, characterized in that, The analysis of the user's historical external interface data is performed as follows: The user's historical external interface data includes weather feature data, road congestion index, and traffic violation record feature data corresponding to the historical driving route; Data of different dimensions are converted into standardized values ​​to obtain weather characteristic risk values, road congestion risk values, and route traffic accident risk values. A random forest algorithm is used, with the risk values ​​of route traffic accidents that have occurred on historical routes as labels, to train the standardized weather characteristic risk values, road congestion risk values, and route traffic accident risk values, and output the route risk index. This establishes a route risk index model that takes weather characteristic risk values ​​and road congestion risk values ​​as input and outputs the route risk index.

5. The safety risk assessment and analysis system based on a minivan according to claim 1, characterized in that, The specific process for obtaining the user's current vehicle aging index, current aggressive driving index, and current route risk index is as follows: The user's current vehicle driving data, current driving behavior data, and current external interface data are respectively input into the vehicle body aging index risk model, the driving aggression index risk model, and the route risk index model to obtain the user's current vehicle body aging index, current driving aggression index, and current route risk index. The vehicle driving data, driving behavior data and external interface data collected by the user in each instance in the database are input into the vehicle body aging index risk model, the driving aggression index risk model and the route risk index model, respectively, to obtain the vehicle body aging index, driving aggression index and route risk index collected by the user in each instance. The threshold values ​​for vehicle aging index, driving aggression index, and route risk index are obtained from the user's vehicle aging index, driving aggression index, and route risk index collected from each statistical analysis. Based on the user's current vehicle aging index, current driving aggression index, and current route risk index, the user's basic risk index and personal risk correction factor are derived.

6. The safety risk assessment and analysis system based on a minivan according to claim 5, characterized in that, The analysis yields the user's risk level, a risk heatmap for the region, and a trend analysis report for the region. The specific analysis process is as follows: The user's base risk index is multiplied by the personal risk adjustment factor to obtain the user's risk index. The risk index range corresponding to each risk level is obtained from the database. If the user's risk index belongs to a certain risk index range, it indicates that the user's risk level is that risk level. The risk level and location of each user in the area are obtained by GPS positioning. The risk level of each vehicle is matched with different colors, and the data is marked and aggregated on the electronic map to form a heat map of risk distribution in the area. Based on the user's risk adjustment factor, the risk adjustment factor of each user in the region is obtained. Users whose personal risk adjustment factor is greater than the preset risk adjustment factor are recorded as effective trend users. Based on the characteristic parameters of each violation behavior of each effective trend user in the region, the type of each violation behavior of each effective trend user in the region is obtained through feature recognition technology. The occurrence frequency of each type of violation in the region is counted, thereby obtaining the occurrence frequency of each type of violation behavior in each statistical time period in the region. A coordinate system is established to obtain the trend graph of the change of each type of violation behavior over time, thereby obtaining a trend analysis report.

7. The safety risk assessment and analysis system based on a minivan according to claim 6, characterized in that, The user alert process is as follows: When a user's risk level is higher than the basic risk level, the user will be notified via SMS. When a user's risk level is higher than the standard risk level, the user will be notified via SMS and an audible and visual alarm will be triggered.

8. The safety risk assessment and analysis system based on a minivan according to claim 7, characterized in that, The analysis yielded the risk area distribution and high-risk driving behavior sequences. The specific analysis process is as follows: The region is divided into grid units according to a preset scale. Each grid serves as the basic unit for spatial analysis. The risk level information of all vehicles within the grid is recorded. The number and proportion of vehicles of each risk level within each grid are counted. A spatial clustering algorithm is used to aggregate grids with a density exceeding a threshold for each risk level to form continuous risk areas of each risk level. This yields the boundaries of risk areas for each risk level and the distribution of risk areas. By using image recognition technology, the slope of various violations is obtained from the trend graph of various violations over time. The violations are then sorted in descending order of slope to obtain a sequence of high-risk driving behaviors.

9. The safety risk assessment and analysis system based on a minivan according to claim 8, characterized in that, The specific generation process of the generated regional supervision scheme is as follows: Regarding the distribution of risk areas: temporary law enforcement points and fixed monitoring equipment will be added in risk areas with a risk level higher than the standard risk level. The priority of each risk area in the patrol route will be increased according to the risk level of each risk area. For high-risk driving behavior sequences: a special rectification campaign will be launched, and each type of violation at the top of the sequence will be recorded as a high-risk driving behavior. Accident cases of high-risk driving behaviors will be sent to enterprises in the region to warn them to strengthen driver training. In addition, regional violation warning announcements will be issued in conjunction with traffic management departments to increase the monitoring density of corresponding violation types.

Images