Methods for determining the overall risk of electric bicycle riders, computer-readable storage media and electronic devices

Abstract

This invention discloses a method for determining the comprehensive risk of electric bicycle riders, a computer-readable storage medium, and an electronic device. The comprehensive risk determination method includes: S1 collecting pre-accident risk significance index data for different helmets; S2 collecting post-accident accident severity index data for different helmets; S3 determining the weight of each index based on the index data collected in steps S1 and S2, and constructing a comprehensive risk assessment model; S4 determining the comprehensive risk of riders wearing different helmets according to the comprehensive risk assessment model. This risk assessment takes into account both proactive prediction and passive protection dimensions, achieving accurate evaluation of the rider's comprehensive risk and providing a scientific basis for helmet product selection, safety standard formulation, and cycling safety management. The risk assessment is comprehensive, reliable, and highly practical.

Description

Technical Field

[0001] This invention belongs to the field of traffic safety assessment technology, specifically relating to a method for determining the comprehensive risk of electric bicycle riders, a computer-readable storage medium, and an electronic device. Background Technology

[0002] With the acceleration of urbanization and the popularization of green travel concepts in my country, electric bicycles have become an important short-distance transportation tool. According to statistics from the Ministry of Public Security, as of July 2025, the number of electric bicycles in China had exceeded 380 million. As a result, non-motorized vehicle traffic accidents have become frequent, with traumatic brain injury being the main cause of injury, disability, and even death for riders.

[0003] Studies have shown that wearing a safety helmet correctly can effectively reduce the risk of head injury by approximately 70% and the mortality rate by 40%. Therefore, since 2020, a nationwide "One Helmet, One Belt" safety campaign has been implemented, mandating that electric bicycle riders wear safety helmets, significantly raising public awareness. However, current research mainly focuses on "whether to wear a helmet," neglecting the issue of "what type of helmet to wear."

[0004] Currently, common electric bicycle safety helmets can be divided into three categories according to their structure: full-face helmets, 3 / 4 helmets, and half helmets. Each type of helmet differs significantly in terms of protection range, field of vision, ventilation, and wearing comfort: Full-face helmets: cover the entire head, offering the best protection, but are heavier, have limited field of vision, and poor ventilation; 3 / 4 helmets: cover the top, sides, and back of the head, leaving the face open, balancing protection and comfort; Half helmets: protect only the upper part of the head, are lightweight but offer the smallest protective area.

[0005] Existing research mainly focuses on two directions: one is to evaluate the energy absorption capacity and shock absorption effect of different helmets after an accident through crash tests or simulations; the other is to detect whether a cyclist is wearing a helmet based on image recognition algorithms. However, few studies systematically explore the impact mechanism of helmet type on a cyclist's emergency response capability before an accident, and there is a lack of methods to integrate pre-accident reaction time and post-accident protective effectiveness into a comprehensive evaluation system. This makes it impossible to scientifically determine the comprehensive safety risk of cyclists corresponding to different helmet types, and it is difficult to provide comprehensive support for helmet selection, product design optimization, and cycling safety management.

[0006] Chinese patent document CN114997714A discloses a risk identification method, system, and computer equipment for electric bicycle riders. It focuses solely on "pre-accident violation risk," using traffic violation information as core data, without addressing post-accident injury protection effectiveness assessment. This constitutes a "single pre-accident prediction," and lacks a clear weight calculation method, relying solely on fuzzy clustering algorithms for risk level classification. The weights depend on the algorithm's default logic and lack specificity. Chinese patent document CN120096725A discloses an intelligent control system and method for electric bicycles. It focuses solely on "real-time driving risk," monitoring the driving environment, vehicle status, and riding behavior, without considering the impact of helmet type on risk or covering post-accident protection dimensions. This constitutes "single process monitoring." Furthermore, it uses preset judgment conditions (such as speed thresholds and angle thresholds) for risk assessment, without quantitative weight allocation, resulting in strong subjectivity and failing to reflect differences in the importance of indicators. Summary of the Invention

[0007] To address the aforementioned technical problems, this invention provides a method for determining the comprehensive risk of electric bicycle riders, a computer-readable storage medium, and an electronic device. The risk assessment takes into account both proactive prediction and passive protection dimensions, achieving accurate evaluation of the rider's comprehensive risk. Furthermore, it provides a scientific basis for helmet product selection, safety standard formulation, and cycling safety management. The risk assessment is comprehensive, the results are reliable, and it is highly practical.

[0008] To achieve the above-mentioned technical objectives, the technical solution adopted by this invention is: a method for determining the comprehensive risk of electric bicycle riders, comprising the following steps: S1: Collect data on the pre-accident risk significance indicators for different helmets; S2: Collect post-accident severity index data for different helmets; S3: Based on the indicator data collected in steps S1 and S2, determine the weight of each indicator and construct a comprehensive risk assessment model. S4: Determine the overall risk of riders wearing different helmets based on the comprehensive risk assessment model.

[0009] This invention collects data on core indicators of pre-accident risk salience and post-accident severity, determines the weight of each indicator, and then constructs a comprehensive risk assessment model. Based on this model, it assesses the overall risk of cyclists wearing different helmets. For the first time, it incorporates the impact of helmets on cyclists' reaction time into the risk assessment system, overcoming the shortcomings of traditional assessments that only focus on "passive protection." It achieves a transformation from "static protection" to an integrated approach of "dynamic response + static protection," enabling accurate assessment of cyclists' overall risk. Furthermore, it provides a scientific basis for helmet product selection, safety standard setting, and cycling safety management. The risk assessment is comprehensive, reliable, and highly practical.

[0010] Furthermore, the pre-accident risk significance indicator in step S1 is the pre-accident emergency response time, and the post-accident accident severity indicator in step S2 is the post-accident head and neck injury risk.

[0011] Furthermore, the data on the emergency response time before the accident is obtained through real vehicle testing. The method is to conduct real vehicle testing in a closed test scenario and record the reaction time of the rider from observing the obstacle to moving their hand to braking.

[0012] Furthermore, the index data of head and neck injury risk after the accident were obtained through bibliometrics. The method involved screening empirical research literature, extracting head and neck injury data corresponding to different types of helmets, statistically analyzing the probability of head and neck injury risk, and normalizing the data.

[0013] Furthermore, in step S3, the determination of the weights of each indicator adopts a method combining objective weights and subjective weights to obtain a comprehensive weight.

[0014] Furthermore, the method for combining the objective weights and subjective weights is a multiplicative synthesis method; the objective weights are calculated based on the degree of data dispersion, and the subjective weights are obtained through expert scoring.

[0015] By integrating objective and subjective weights through multiplication and synthesis, the overall weights become more dependent on the consistent contributions of both subjective and objective weights. This avoids extreme biases of a single weight and ensures the scientific and reliable nature of weight assignment.

[0016] Furthermore, the comprehensive risk assessment model in step S3 is a linear weighted assessment model; the expression of the comprehensive risk assessment model is: , where A i Let h1 be the comprehensive risk value of the i-th type of helmet, h1 be the comprehensive weight of the emergency response time before the accident, and t be the comprehensive risk value of the i-th type of helmet. i Let h1 be the normalized pre-accident emergency response time for the i-th type of helmet, h2 be the comprehensive weight of the risk of head and neck injury after the accident, and m be the total head and neck injury risk. i The normalized risk of head and neck injury after an accident for the i-th type of helmet.

[0017] Furthermore, in step S4, when determining the overall risk of a cyclist, the risk level is determined by comparing it with a preset risk threshold; the risk level includes at least low risk, medium risk, and high risk.

[0018] The present invention also provides a computer-readable storage medium on which a computer program running on it can assess the comprehensive risk of cyclists wearing different helmets. The risk assessment takes into account both active prediction and passive protection, achieving accurate assessment of the comprehensive risk of cyclists. It also provides a scientific basis for helmet product selection, safety standard setting and cycling safety management. The risk assessment is comprehensive, reliable and practical.

[0019] To achieve this technical objective, the present invention employs the following technical solution: a computer-readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to perform any of the methods described above.

[0020] The present invention also provides an electronic device that can assess the comprehensive risk of cyclists wearing different helmets. Its risk assessment takes into account both active prediction and passive protection, achieving accurate assessment of the comprehensive risk of cyclists. It also provides a scientific basis for helmet product selection, safety standard setting and cycling safety management. The risk assessment is comprehensive, reliable and practical.

[0021] To achieve this technical objective, the present invention employs the following technical solution: the electronic device includes a memory and a processor, wherein the memory is used to store a program that supports the processor in executing any of the methods described above, and the processor is configured to execute the program stored in the memory. Attached Figure Description

[0022] The following detailed description, in conjunction with the accompanying drawings and embodiments of the present invention, is as follows: Figure 1 This is a bar chart comparing objective data on emergency response time before an accident and the risk of head and neck injury after an accident for different helmets of this invention; Figure 2 This is a comparative analysis chart of the objective weight, subjective weight, and comprehensive weight of the core indicators of emergency response time before an accident and the risk of head and neck injury after an accident, as presented in this invention. Detailed Implementation

[0023] To enhance understanding of the present invention, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. These embodiments are only used to explain the invention and do not limit the scope of protection of the invention.

[0024] The method for determining the overall risk of electric bicycle riders provided in this embodiment includes the following steps: S1: Collect data on the pre-accident risk significance indicators for different helmets; S2: Collect post-accident severity index data for different helmets; S3: Based on the indicator data collected in steps S1 and S2, determine the weight of each indicator and construct a comprehensive risk assessment model. S4: Determine the overall risk of riders wearing different helmets based on the comprehensive risk assessment model.

[0025] This invention collects data on core indicators of pre-accident risk salience and post-accident severity, determines the weight of each indicator, and then constructs a comprehensive risk assessment model. Based on this model, it assesses the overall risk of cyclists wearing different helmets. For the first time, it incorporates the impact of helmets on cyclists' reaction time into the risk assessment system, overcoming the shortcomings of traditional assessments that only focus on "passive protection." It achieves a transformation from "static protection" to an integrated approach of "dynamic response + static protection," enabling accurate assessment of cyclists' overall risk. Furthermore, it provides a scientific basis for helmet product selection, safety standard setting, and cycling safety management. The risk assessment is comprehensive, reliable, and highly practical.

[0026] The pre-accident risk significance indicator in step S1 is the pre-accident emergency response time, and the post-accident accident severity indicator in step S2 is the risk of head and neck injury after the accident.

[0027] The aforementioned pre-accident emergency response time is a complex response time, covering the entire process of the cyclist's visual recognition, decision-making, and action execution.

[0028] The helmets are classified into three categories based on their structural design: full-face helmets, 3 / 4 helmets, and half helmets, which serve as the basic classification standard for subsequent evaluation. The full-face, 3 / 4, and half helmets selected meet legal safety standards, ensuring that all three types comply with the "Motorcycle and Electric Bicycle Passenger Helmets" safety standard.

[0029] This application constructs a two-dimensional assessment system of "pre-accident risk salience + post-accident severity". The pre-accident dimension selects "reaction time" as the core indicator, directly reflecting the helmet's impact on the rider's emergency response capability; the post-accident dimension selects "head and neck injury risk" as the core indicator, quantifying the helmet's passive protective effectiveness. Furthermore, helmet type is identified as the core influencing variable, and the application systematically analyzes the differentiated performance of three mainstream helmet types—full-face helmets, 3 / 4 helmets, and half-face helmets—under both dimensions.

[0030] In step S3, the determination of the weight of each indicator adopts a method combining objective weight and subjective weight to obtain a comprehensive weight; the objective weight is calculated based on the data dispersion, and the subjective weight is obtained through expert scoring.

[0031] In this embodiment, the weights are determined by combining the entropy weight method with the Likert scale expert rating method. The entropy weight method calculates objective weights based on data dispersion; the Likert scale method invites at least 10 experts in relevant fields to conduct subjective ratings, and calculates subjective weights based on the average score and coefficient of variation. The Likert scale rating levels include very important, important, moderately important, not very important, and unimportant, corresponding to scores of 5, 4, 3, 2, and 1 points, respectively. In other embodiments, the objective weights can also be obtained using methods such as the coefficient of variation method, the CRITIC method, and principal component analysis; the subjective weights can also be obtained using methods such as the Delphi method, the analytic hierarchy process, and the pairwise comparison method.

[0032] The objective data on emergency response time before an accident was obtained through real-vehicle testing. The method was as follows: Real-vehicle testing was conducted in a closed test scenario (closed road section), setting up two complex traffic scenarios: left turn and right turn. The closed test scenario included road sections with invisible obstacles and intersections with poor visibility. The visible point was changed by moving the position of the obstacle to eliminate interference from road conditions, weather, and psychological factors. During the test, the rider's fingers did not initially touch the brake. Using a blue triangular cone as the obstacle, the time from when the rider discovered the obstacle (the visible point of the obstacle) to when their hand moved to the brake was recorded. After video analysis, outlier removal, and normality testing, the effective reaction time data was obtained.

[0033] Specifically, the actual vehicle testing process is as follows: First, the test subjects, wearing different helmets, rode on sections of road where the obstacle was not visible. Then, at intersections with poor visibility, they prepared to turn left or right. Due to the obstruction of buildings, the blue triangular cone obstacle was only visible to the test subjects when they reached a certain point near the intersection; this point was called the obstacle visibility point. During the test, the time from when the test subject rode to the obstacle visibility point to when they moved their hand to the brake was the electric bicycle's reaction time. During the test, the position of the obstacle would move back and forth, and the position of the obstacle visibility point would also change accordingly to ensure the reliability of the data. The collected data on emergency response time before the accident (objective data) were then processed. The method was as follows: the video was analyzed frame by frame to calculate the time from when the cyclist's hand moved to the brake from the point of visibility of the obstacle, and the unit was converted to milliseconds. Outliers were removed using the Puata criterion, and the remaining data were tested for normality to confirm that the data conformed to a normal distribution.

[0034] In this embodiment, reaction times were collected 212 times, and the pre-accident emergency reaction time index data (objective data) for different helmets were statistically obtained as follows: full-face helmet: left turn 960.61 ms, right turn 973.68 ms, average 967.15 ms; 3 / 4 helmet: left turn 734.17 ms, right turn 773.15 ms, average 753.66 ms; half helmet: left turn 672.50 ms, right turn 647.22 ms, average 659.86 ms. The data were processed using proportional normalization, and the calculation formula is as follows: ,in, This represents the result after normalization. This represents the raw reaction time data corresponding to a specific helmet. These are the raw reaction time data for the i-th helmet. The results are: full-face helmet 0.40, 3 / 4 helmet 0.32, and half helmet 0.28.

[0035] The objective data on the risk of head and neck injury after an accident were obtained through bibliometrics. The method was as follows: relevant empirical research literature was retrieved from literature databases. The screening criteria included: the research subjects were electric bicycle or motorcycle riders, the data included comparative data on full-face helmets, 3 / 4 helmets, and half helmets, the outcome indicators of head and neck injury (such as traumatic brain injury, facial fractures, etc.) were clearly defined, the sample size was not less than 100 cases, the head and neck injury case data corresponding to different types of helmets were extracted, and the probability of head and neck injury risk was statistically analyzed. The probability of head and neck injury risk was normalized to obtain a dimensionless head and neck injury risk value.

[0036] In this embodiment, head and neck injury risk data (objective data) were obtained by searching review articles in the PubMed, Scopus, and Web of Science databases using the keywords "helmet" and "head and neck injury." Six articles meeting the requirements were ultimately selected, covering 6529 participants. The number of head and neck injuries corresponding to different helmet types was statistically analyzed. The conclusion was that full-face helmets were significantly more effective than half-face and 3 / 4 helmets in preventing head and neck injuries, with a reduction ratio of 0.356 relative to half-face helmets and 0.636 relative to 3 / 4 helmets. After proportional normalization, the injury risk data were: full-face helmet 0.19, 3 / 4 helmet 0.29, and half-face helmet 0.52.

[0037] The objective data on the emergency response time before the accident and the risk of head and neck injury after the accident are shown in Table 1. A comparative analysis of these data yielded the following results. Figure 1 .

[0038] Table 1. Objective data on emergency response time before an accident and the risk of head and neck injury after an accident for different helmets.

[0039] The specific method for determining the indicator weights in step S3 is as follows: The entropy weight method calculates objective weights and standardizes objective data on emergency response time before an accident and risk of head and neck injury after an accident, eliminating the influence of dimensions.

[0040] The calculation process is described below, and the calculation results are shown in Table 2: The first step is to turn negative indicators positive (eliminate the difference in indicator direction).

[0041] Since both metrics are "smaller values ​​indicate better performance", they need to be converted into positive metrics where "larger values ​​indicate better performance". The formula is as follows: ,in, For the j-th evaluation index, the original index data for the i-th type of helmet is... The maximum value of the j-th indicator is used to ensure that the conversion logic is unique and verifiable. Let be the index value of the i-th type of helmet after the negative value is turned into a positive value under the j-th evaluation index.

[0042] The second step is to calculate the indicator weight ratio P. ij (Avoid zero-value interference).

[0043] To prevent the logarithm from becoming meaningless due to zero values ​​after normalization, a minimal correction coefficient is introduced. It is more accurate than the conventional ε=0.0001, avoiding correction interference. Formula: ,in, Let be the corrected normalized proportion of the i-th type of helmet under the j-th evaluation index. The value is a minimal positive correction number, and n is the number of helmet types. The text mentions full-face helmets, 3 / 4 helmets, and half helmets, so n=3.

[0044] The third step is to calculate the information entropy. .

[0045] Information entropy reflects the degree of dispersion of data, and is a measure of dispersion. The smaller the entropy, the higher the data dispersion, and the stronger the distinguishing ability of the indicator. The formula is: ,in, Let be the information entropy of the j-th evaluation index.

[0046] Step 4: Calculate the coefficient of difference With objective weight .

[0047] The coefficient of difference reflects the ability of an indicator to distinguish evaluation results. The larger the coefficient of difference, the stronger the distinguishing ability. The formula for the coefficient of difference is: ,in, Let be the difference coefficient of the j-th evaluation index.

[0048] The third step calculates the information entropy of the j-th evaluation indicator. The difference coefficient is then proportionally normalized to obtain the objective weight. The sum of these weights is 1, objectively reflecting the proportion of indicator importance. The formula for calculating the objective weight is: ,in, q represents the objective weight of the j-th evaluation indicator, which is the weight calculated by the entropy weight method; q represents the number of evaluation indicators. The paper involves two indicators: emergency response time before the accident and risk of head and neck injury after the accident, so q=2.

[0049] Table 2. Relevant data on different influencing factors calculated using the entropy weight method.

[0050] Subjective weighting was obtained through expert scoring using the Likert scale: 10 experts were invited to score the questionnaire; the questionnaire referenced a 5-point Likert scale, with each question having 5 options, such as "How important do you think reaction time is to helmet performance?": Very important, Somewhat important, Generally important, Not very important, Not important, Not important; The question selected two key indicators: reaction time (reflecting riding maneuverability) and head and neck injury risk (reflecting accident protection safety). The calculation process is described below, and the results are shown in Table 3.

[0051] The first step is to calculate the statistical scores for each indicator.

[0052] The average score reflects the average perceived importance of the indicator, and its calculation formula is as follows: ,in, M is the mean score of the j-th evaluation. ij Let be the score given by the i-th expert to the j-th indicator, where i is the expert's index, n is the number of experts, and j is the index of the evaluation indicator; the standard deviation reflects the dispersion of experts' perception of the importance of the indicator, and its calculation formula is: ,in, Let be the standard deviation of the j-th evaluation indicator; the coefficient of variation is the standardized dispersion, which can eliminate the influence of the score mean, and its calculation formula is: ,in, Let be the coefficient of variation of the j-th evaluation index.

[0053] The second step is to calculate the "importance correction coefficient" for the indicator.

[0054] A smaller coefficient of variation indicates a greater consensus among experts regarding the importance of the indicator, thus increasing the reliability of the indicator's score and requiring a higher correction weight. Correction coefficient formula: ,in, Let be the importance correction coefficient for the j-th evaluation indicator. Let be the "consistency score" of the j-th indicator, and m be the number of evaluation indicators.

[0055] The third step is to calculate the subjective weights.

[0056] Combining the mean score of the indicators with the correction factor, the final weight is "standardized mean × correction factor", ensuring that the weight reflects both average importance and cognitive consistency. The formula for calculating the standardized mean of the indicators is as follows: ,in, Let be the average score of the j-th indicator; the formula for calculating the subjective weight is: ,in, Let be the subjective weight of the j-th evaluation indicator, which is the weight calculated using the expert scoring method of the Likert scale. Let be the standardized value of the mean score of the j-th indicator.

[0057] Table 3. Relevant data on different influencing factors calculated using the Likert scale.

[0058] In summary, in this embodiment, the entropy weight method yielded the following results: reaction time information entropy 0.612598, coefficient of variation 0.387402, weight 50.24%; head and neck injury risk information entropy 0.616198, coefficient of variation 0.383802, weight 49.76%; and Likert scale expert scores yielded the following results: average reaction time score 3.4, coefficient of variation 0.2056, weight 39.1%; and average head and neck injury risk score 4.7, coefficient of variation 0.1028, weight 60.9%.

[0059] The method for combining objective weights and subjective weights is the multiplicative composition method, and the calculation formula for the multiplicative composition method is as follows: , where w j The comprehensive weight of the j-th indicator is... Let j be the objective weight of the j-th indicator. Let be the subjective weight of the j-th indicator, n be the total number of evaluation indicators, and k be the index of the indicator, from 1 to n, used to iterate through all indicators and calculate the sum of the denominators.

[0060] By substituting the objective and subjective weights obtained above into the multiplicative synthesis formula, the comprehensive weights of the emergency response time before the accident and the risk of head and neck injury after the accident can be calculated.

[0061] In this embodiment, the overall weight of the emergency response time before the incident is 61.21%, and the overall weight of the risk of head and neck injury after the incident is 38.79%.

[0062] The above-mentioned comparative analysis of the objective weight, subjective weight, and comprehensive weight of emergency response time and core indicators of head and neck injury risk after an accident is shown in the figure below. Figure 2 As shown.

[0063] This invention employs the core logic of the entropy weight method combined with a Likert scale to calculate the comprehensive weight. The entropy weight method (objective weight) calculates weights based on the actual data dispersion of reaction time and head and neck injury risk, avoiding human intervention. The calculation logic reflects data dispersion through information entropy and calculates weights through the coefficient of difference, ensuring a direct correlation between weights and data characteristics. The Likert scale (subjective weight) involves inviting 10 experts in relevant fields to score, calculating weights based on the importance level of indicators (out of 5), incorporating industry experience. Finally, the comprehensive weight is calculated by merging subjective and objective weights, strengthening consensus and avoiding extreme biases from single weights. Through this fusion of subjective and objective weights, the weight allocation combines "data objectivity" and "empirical rationality," resulting in more reliable results. It also clarifies the importance proportions of the two-dimensional indicators, providing a scientific basis for subsequent risk model construction.

[0064] Because the importance of each influencing factor to the evaluation objective is heterogeneous, their corresponding weight allocations show significant differences. The comprehensive risk assessment model in step S3 is constructed based on the linear weighted summation method. Specifically, this linear weighted assessment model uses reaction time and head and neck injury risk as input variables, combined with comprehensive weights, to calculate the comprehensive risk value corresponding to different types of helmets.

[0065] A comprehensive risk assessment model is constructed based on a linear weighted model, and its expression is: , where A i Let t be the overall risk value of the i-th type of helmet. i m is the normalized value of the reaction time. i This is a normalized value for the risk of head and neck injuries. In this invention, the expression for the comprehensive risk assessment model is: , where A i Let h1 be the comprehensive risk value of the i-th type of helmet, h1 be the comprehensive weight of the emergency response time before the accident, and t be the comprehensive risk value of the i-th type of helmet. i Let h1 be the normalized pre-accident emergency response time for the i-th type of helmet, h2 be the comprehensive weight of the risk of head and neck injury after the accident, and m be the total head and neck injury risk. i The normalized risk of head and neck injury after an accident for the i-th type of helmet.

[0066] In this embodiment, the comprehensive risk value is calculated using a comprehensive risk assessment model and is as follows: full-face helmet A1≈0.32; 3 / 4 helmet A2≈0.30; half helmet A3≈0.37.

[0067] In step S4, when assessing the overall risk of a cyclist, the risk level is determined by comparing it with a preset risk threshold; the risk level includes at least low risk, medium risk, and high risk.

[0068] The preset risk thresholds include a first threshold of 0.30 and a second threshold of 0.35. The specific quantitative risk value calculated by the model is the core, and the risk range and level of the helmet are defined by comparing the relative magnitudes of the values. The lower the value, the lower the overall risk. Simultaneously, considering the core objective of traffic risk prevention and control—safety priority over response efficiency—the practical application of the risk range is further defined. Risk levels are divided according to the magnitude of the overall risk value: an overall risk value ≤ 0.30 is low risk, 0.30 < overall risk value ≤ 0.35 is medium risk, and an overall risk value > 0.35 is high risk. This is used to determine the overall risk level of riders under different helmet types.

[0069] In this embodiment, according to the risk level determination criteria, full-face helmets are considered medium risk, 3 / 4 helmets are considered low risk, and half-face helmets are considered high risk. That is, 3 / 4 helmets have the lowest overall risk, followed by full-face helmets, and half-face helmets have the highest risk.

[0070] This invention provides a comprehensive risk assessment covering the entire chain from "risk prediction to accident prevention," avoiding the one-sidedness of a single-dimensional assessment; it can directly provide quantitative basis for cyclists to select helmets and optimize helmet product design.

[0071] This embodiment also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, performs the above-described method for determining the comprehensive risk of electric bicycle riders.

[0072] This embodiment also provides an electronic device, including a memory and a processor. The memory is used to store a program that supports the processor in executing the above-described method for determining the comprehensive risk of electric bicycle riders. The processor is configured to execute the program stored in the memory.

[0073] This invention proposes a dual-dimensional assessment framework linking pre- and post-event responses, and for the first time incorporates the impact of helmets on rider reaction time into the risk assessment system. This overcomes the shortcomings of traditional assessments that only focus on "dynamic protection," achieving a shift from "static protection" to an integrated approach of "dynamic response + static protection." The quantitative indicators are scientifically integrated, the data sources are diverse and reliable, and expert consensus verification is incorporated to ensure the model possesses good reliability and validity, making it easily adoptable by the industry.

[0074] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for determining the comprehensive risk of electric bicycle riders, characterized in that, Includes the following steps: S1: Collect data on the pre-accident risk significance indicators for different helmets; S2: Collect post-accident severity index data for different helmets; S3: Based on the indicator data collected in steps S1 and S2, determine the weight of each indicator and construct a comprehensive risk assessment model. S4: Determine the overall risk of riders wearing different helmets based on the comprehensive risk assessment model.

2. The method for determining the comprehensive risk of electric bicycle riders according to claim 1, characterized in that, The pre-accident risk significance indicator in step S1 is the pre-accident emergency response time, and the post-accident accident severity indicator in step S2 is the risk of head and neck injury after the accident.

3. The method for determining the comprehensive risk of electric bicycle riders according to claim 2, characterized in that, The data on the emergency response time before the accident was obtained through real vehicle testing. The method involved conducting real vehicle testing in a closed test scenario and recording the rider's reaction time from observing the obstacle to moving their hand to braking.

4. The method for determining the comprehensive risk of electric bicycle riders according to claim 2, characterized in that, The data on the risk of head and neck injury after an accident were obtained through bibliometrics. The method involved screening empirical research literature, extracting head and neck injury data corresponding to different types of helmets, statistically analyzing the probability of head and neck injury risk, and then normalizing the data.

5. The method for determining the comprehensive risk of electric bicycle riders according to claim 2, characterized in that, In step S3, the determination of the weight of each indicator adopts a method that combines objective weight and subjective weight to obtain a comprehensive weight.

6. The method for determining the comprehensive risk of electric bicycle riders according to claim 5, characterized in that, The method for combining objective weights and subjective weights is a multiplicative synthesis method; the objective weights are calculated based on the degree of data dispersion, and the subjective weights are obtained through expert scoring.

7. The method for determining the comprehensive risk of electric bicycle riders according to claim 5, characterized in that, The comprehensive risk assessment model in step S3 is a linear weighted assessment model; the expression of the comprehensive risk assessment model is: , where A i Let h1 be the comprehensive risk value of the i-th type of helmet, h1 be the comprehensive weight of the emergency response time before the accident, and t be the comprehensive risk value of the i-th type of helmet. i Let h1 be the normalized pre-accident emergency response time for the i-th type of helmet, h2 be the comprehensive weight of the risk of head and neck injury after the accident, and m be the total head and neck injury risk. i The normalized risk of head and neck injury after an accident for the i-th type of helmet.

8. The method for determining the comprehensive risk of electric bicycle riders according to claim 1, characterized in that, In step S4, when assessing the overall risk of a cyclist, the risk level is determined by comparing it with a preset risk threshold. The risk levels include at least low risk, medium risk, and high risk.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, The computer program is executed by the processor to perform the method described in any one of claims 1-8.

10. An electronic device, comprising a memory and a processor, characterized in that, The memory is used to store a program that supports a processor in executing the method of any one of claims 1-8, the processor being configured to execute the program stored in the memory.

Images