Meteorological disaster-oriented primary safety production accident risk assessment method and system

Abstract

The application relates to the technical field of data processing, and discloses a grass-roots safety production accident risk assessment method and system for meteorological disaster factors, which comprises the following steps: collecting historical meteorological data, safety production accident data, urbanization rate data and total rural population; performing data preprocessing on the collected historical meteorological data, safety production accident data, urbanization rate data and total rural population; performing data classification on the preprocessed historical meteorological data and safety production accident data; adopting a two-way fixed effect model as a benchmark model to analyze the influence of meteorology on the frequency and severity of grass-roots safety production accidents; adding the urbanization rate data and total rural population as adjustment variables into the benchmark model to establish a grass-roots safety production accident risk assessment model. The application expands the current understanding of weather factors and provides scientific guidance for grass-roots safety supervision and individual prevention.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, specifically to a method and system for risk assessment of grassroots work safety accidents oriented towards meteorological disaster-causing factors. Background Technology

[0002] Work safety accidents not only cause direct physical and mental harm to employees but also negatively impact business performance. Recent reports show that the overall number of work safety accidents in major global economies is declining. Despite this positive trend, significant differences remain in the spatial distribution of work safety accidents. Rural areas are highly dependent on outdoor, weather-sensitive industries such as agriculture, forestry, and construction, and face challenges such as weak infrastructure, limited safety supervision, and slow emergency response. However, with urbanization, the industrial structure of many rural areas is becoming increasingly diversified, adding manufacturing, logistics, and service activities beyond traditional agriculture. This transformation alters the types and patterns of work safety accidents, but local safety management systems often remain inadequate. These combined characteristics make rural workers particularly vulnerable to weather-related occupational risks.

[0003] At the same time, emerging environmental challenges are reshaping occupational risks. The ongoing impact of meteorological conditions, such as rising temperatures, frequent extreme weather events, and increased climate variability, is constantly altering the work environment and the probability of accidents. Meteorological conditions should be considered a core element of occupational safety research.

[0004] Existing research has examined the relationship between workplace accidents and extreme weather events. For example, Bendak et al. (2022) studied the impact of high ambient temperatures on worker performance. However, these studies mostly focus on a single meteorological dimension, rarely considering the combined or interactive effects of multiple meteorological factors, resulting in an incomplete understanding of their comprehensive impact. Furthermore, with the urbanization and economic diversification in rural areas (factory establishment, infrastructure expansion, and non-agricultural industry transformation), occupational hazards are becoming increasingly complex and diverse, but existing research is still largely limited to agricultural scenarios. Therefore, we have developed a grassroots workplace accident risk assessment method oriented towards meteorological disaster-causing factors, solving the above-mentioned technical problems. Summary of the Invention

[0005] This invention provides a method and system for risk assessment of grassroots work safety accidents oriented towards meteorological disaster factors. By quantifying the impact of various weather factors on accidents and introducing urbanization rate and total rural population as moderating variables, a more accurate risk assessment model is constructed, thereby providing scientific guidance for grassroots safety supervision and individual prevention.

[0006] Therefore, the present invention provides the following technical solution: A method for risk assessment of workplace safety accidents at the grassroots level, oriented towards meteorological disaster-causing factors, the method comprising: Step 1: Collect historical meteorological data, work safety accident data, urbanization rate data, and total rural population. Step 2: Perform data preprocessing on the collected historical meteorological data, work safety accident data, urbanization rate data, and total rural population. Step 3: Classify the preprocessed historical meteorological data and the work safety accident data; Step 4: Use a two-way fixed effects model as the baseline model to analyze the impact of meteorology on the frequency and severity of grassroots work safety accidents; Step 5: Add the urbanization rate data and the total rural population as adjustment variables to the baseline model to establish a grassroots work safety accident risk assessment model.

[0007] Optionally, in step 1, the historical meteorological data includes temperature data, humidity data, atmospheric pressure data, wind speed data, cloud cover data, rainfall data, snowfall data, and fog concentration data; the work safety accident data includes the accident occurrence time, accident location, accident frequency, accident cause, number of deaths, and number of injuries; accidents are classified according to their causes; accidents are divided into three levels according to the number of deaths and injuries: accidents with fewer than 3 deaths or fewer than 10 injuries are classified as general accidents; accidents causing 3 to 10 deaths or 10 to 50 serious injuries are classified as major accidents; accidents causing more than 10 deaths or more than 50 injuries are classified as serious and extremely serious accidents; the accident frequency is measured by the number of accidents per month. Optionally, in step 1, the historical meteorological data is collected on a monthly basis, and the temperature data includes the monthly average temperature. Temperature it and the square of the monthly average temperature Temperature it ² Cloud data Cloud it The monthly average total cloud cover is calculated as the percentage of the sky obscured by clouds, ranging from 0% (clear sky) to 100% (overcast); rainfall data. Rain it Includes light rain, moderate rain, heavy rain, showers, thunderstorms, and the number of rainy days; snowfall data. Snow it Includes data on light snow, snow, heavy snow, sleet, ice pellets, and the number of days with snowfall; fog concentration data. Fog it Includes light fog, dense fog, heavy fog, and the number of days with fogfall; atmospheric pressure data. Pressure itThe monthly average sea-level pressure, quantified in hectopascals, reflects the monthly atmospheric pressure changes; wind speed data... Windspeed it Monthly average wind speed, measured in meters per second, represents the average airflow velocity at the study location; humidity data includes relative humidity. Humidity it The product of temperature and humidity Humidity it ×Temperature it That is, the humidity-temperature product term is calculated by multiplying the relative humidity by the temperature, where relative humidity is... Humidity it In percentage terms.

[0008] In step 1, the urbanization rate data and total population are collected, and the total rural population is calculated: Total rural population = Total population (1 - Urbanization rate data).

[0009] Step 2 includes: Step 21: Clean the collected historical meteorological data, work safety accident data, urbanization rate data, and total rural population data by standardizing date and character formats, removing duplicate observations, standardizing variable naming, and specifying individual identifiers and time identifiers. Step 22: When the historical meteorological data, the work safety accident data, the urbanization rate data, and the total rural population have missing values ​​after data cleaning, if the missing value is a key variable, the observation row containing the missing value is directly deleted. Step 23: Perform data detection and outlier processing on the historical meteorological data, the work safety accident data, the urbanization rate data, and the total rural population after missing value processing. Use box plots or interquartile ranges to identify extreme values, remove obviously erroneous values, and perform tail reduction or logarithmic transformation on real extreme values ​​to reduce interference with the mean-free estimation within the group. Step 24: Perform variable transformation on the historical meteorological data, work safety accident data, urbanization rate data, and total rural population after outlier processing. If the variable distribution is severely skewed, logarithmic transformation is used. At the same time, check for multicollinearity between interaction terms and original variables, calculate the variance inflation factor (VIF), and if the VIF exceeds 10, perform variable centering or delete redundant terms. Step 25: Verify whether the historical meteorological data has sufficient temporal variation within an individual, i.e., each individual must have at least two observation points at time. If the condition is not met, delete the data. Step 26: Exploratory data analysis is conducted through descriptive statistics, correlation heatmaps, and visualization of within-group variation to ensure that the data quality of the historical meteorological data, the work safety accident data, the urbanization rate data, and the total rural population meets the estimation requirements of the two-way fixed effects model.

[0010] In step 3, the accident frequency, number of casualties, number of general accidents, number of major accidents, and number of serious and major accidents in the work safety accident data are defined as the dependent variables; the monthly average temperature in the historical meteorological data is defined as the dependent variable. Temperature it Monthly average temperature squared term Temperature it ² relative humidity Humidity it Cloud data Cloud it Rainfall data Rain it Snowfall data Snow it Fog concentration data Fog it Defined as explanatory variables; atmospheric pressure data from the historical meteorological data. Pressure it 、 Wind speed data Windspeed it The product of temperature and humidity Humidity it × Temperature it Defined as a control variable.

[0011] In step 4, the baseline model is set as follows: ; In the formula, Y it The dependent variables include accident frequency, number of casualties, number of general accidents, number of major accidents, and number of serious and major accidents; β 0 、β 1 、β 2 、β 3 、β 4 、β 5 、β 6 、β 7 、β 8 、β 9 and β 10 These are the weight coefficients for each variable; λ i For regional fixed effects, νt For month-fixed effects, control for time-invariant regional characteristics and time-specific shocks; ε it This is a random disturbance term.

[0012] In step 5, the grassroots work safety accident risk assessment model includes an urbanization rate impact model and a rural population total impact model. The urbanization rate impact model is as follows: ; in, This refers to taking t Urbanization rate data the median of If the base is , Then the corresponding =1, otherwise 0; β 11 、β 12 and β 13 These are the weight coefficients for each variable; The model for the impact of the total rural population is as follows: ; in, This refers to the total rural population. β 14 、β 15 and β 16 These are the weighting coefficients for each variable.

[0013] A grassroots work safety accident risk assessment system oriented towards meteorological disaster-causing factors, the system comprising: The data acquisition unit collects historical meteorological data, work safety accident data, urbanization rate data, and total rural population. The data preprocessing unit preprocesses the collected historical meteorological data, work safety accident data, urbanization rate data, and total rural population. The data classification unit classifies the preprocessed historical meteorological data and the work safety accident data. The baseline model construction unit uses a two-way fixed effects model as the baseline model to analyze the impact of meteorology on the frequency and severity of grassroots work safety accidents; The accident risk assessment model construction unit incorporates the urbanization rate data and the total rural population as adjustment variables into the baseline model to establish a grassroots work safety accident risk assessment model.

[0014] A computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor of the steps of the method for risk assessment of grassroots work safety accidents oriented towards meteorological disaster-causing factors.

[0015] This invention provides a method and system for risk assessment of grassroots work safety accidents oriented towards meteorological disaster-causing factors. It studies the impact of various weather factors on grassroots work safety accidents, innovatively using urbanization rate and population density as key indicators to moderate the impact of meteorological factors on grassroots work safety accidents. The invention analyzes the moderating role of urbanization rate and grassroots population density in the relationship between weather conditions and accident rates. By simultaneously examining different weather factors and the effects of urbanization rate and grassroots population density, it expands the current understanding of weather factors and can provide valuable guidance for individuals, enterprises, and regulatory agencies. Compared with existing technologies, this invention has the following technical advantages: First, this invention breaks through the limitations of traditional methods that only focus on a single meteorological factor. By simultaneously examining the comprehensive impact of multiple weather factors on grassroots work safety accidents, it expands the understanding of meteorological disaster-causing mechanisms and improves the comprehensiveness of risk assessment. Secondly, the present invention uses a fixed effects model for analysis, and the output risk assessment results can not only provide decision support for regulatory agencies to formulate differentiated emergency plans and optimize resource allocation, but also provide individuals and enterprises in different industries with refined safety operation guidance under different weather conditions, which has good practicality and promotion value. Finally, this invention innovatively introduces urbanization rate and population density as moderating variables, revealing the nonlinear impact of weather factors on accident rate under different levels of socioeconomic development; the study found that urbanization rate and population density have a significant moderating effect on weather risk. 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 embodiments will be briefly described below. Obviously, the drawings described below are merely some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart of a grassroots work safety accident risk assessment method oriented towards meteorological disaster-causing factors in a specific embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of a grassroots work safety accident risk assessment system oriented towards meteorological disaster-causing factors in a specific embodiment of the present invention. Detailed Implementation

[0018] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0019] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. 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 should fall within the scope of protection of the present invention.

[0020] like Figure 1 As shown, Figure 1 This is a flowchart of a grassroots work safety accident risk assessment method oriented towards meteorological disaster-causing factors, according to a specific embodiment of the present invention. The method includes: Step 1: Collect historical meteorological data, work safety accident data, urbanization rate data, and total rural population.

[0021] Historical meteorological data includes temperature, humidity, atmospheric pressure, wind speed, cloud cover, rainfall, snowfall, and fog concentration. Work safety accident data includes the time of occurrence, location, frequency, cause, number of deaths, and number of injuries. Accidents are categorized based on their cause, such as fires, drownings, and gas explosions. Based on the number of deaths and injuries, accidents are classified into three levels: accidents with fewer than 3 deaths or fewer than 10 injuries are classified as general accidents; accidents resulting in 3 to 10 deaths or 10 to 50 serious injuries are classified as major accidents; and accidents resulting in more than 10 deaths or more than 50 injuries are classified as serious and extremely serious accidents. Accident frequency is measured by the monthly accident count for each province.

[0022] In one embodiment, historical meteorological data is collected on a monthly basis, and the temperature data includes the monthly average temperature. Temperature it and the square of the monthly average temperature Temperature it 2 Monthly average temperature squared term Temperature it 2This data can reveal potential nonlinear relationships. Cloud cover data is the monthly average total cloud cover, calculated as the percentage of the sky obscured by clouds, ranging from 0% (clear sky) to 100% (overcast). Rainfall data includes light rain, moderate rain, heavy rain, showers, and thunderstorms. Snowfall data includes light snow, snow, heavy snow, sleet, and hail. Fog concentration data includes light fog, dense fog, and thick fog. The number of days with rain, snow, or fog is recorded as the number of rainy days, snowy days, and foggy days. Atmospheric pressure data is the monthly average sea-level pressure, quantified in hectopascals, reflecting the monthly atmospheric pressure changes. Wind speed data is the monthly average wind speed, measured in meters per second, representing the average airflow velocity at the study location. Humidity data is the humidity-temperature product, calculated by multiplying relative humidity by temperature.

[0023] Collect urbanization rate data and total population data. Calculate the total rural population using the following formula: Total rural population = Total population (1 - Urbanization rate data).

[0024] Step 2 involves preprocessing the collected data. This includes: Step 21: Clean the raw data, standardize the date and character formats, remove duplicate observations, standardize variable naming, and specify individual identifiers (such as region IDs) and time identifiers (such as years).

[0025] Step 22: For missing values, if the missing value is a key variable, directly delete the observation row containing the missing value to ensure the data entering the model is complete.

[0026] Step 23, Detect and handle outliers: Identify extreme values ​​using box plots or interquartile ranges. Remove obviously erroneous values. For genuine extreme values, perform tail reduction or logarithmic transformation to reduce their interference with within-group mean estimation. Obviously erroneous values ​​refer to clearly unreasonable values ​​caused by human or system errors during data collection, transcription, or storage. Examples include: a negative accident count; a date field containing "2022-03-40"; an urbanization rate recorded as 150% (normal range 0–100%); and a regional temperature recorded as 99℃ (far exceeding the reasonable climatic range). Genuine extreme values ​​refer to observations that, while within a reasonable range, significantly deviate from the overall data distribution and are at the tail end of the range. For example: In a certain month of a certain year, a plane crash in a certain place resulted in 132 deaths. The 132 deaths reflect a real risk scenario and should not be simply deleted. However, given that it reflects a real risk and should not be arbitrarily removed, tail reduction was used in the analysis, replacing the observations with the number of deaths greater than the 99th percentile with the value of that quantile.

[0027] Step 24: Perform variable transformation: If the variable distribution is severely skewed, use logarithmic transformation; simultaneously check for multicollinearity between the interaction term and the original variables, and calculate the variance inflation factor (VIF). If the VIF exceeds 10, consider variable centering or removing redundant terms. Severe skewness refers to a variable distribution that is asymmetrical, with the tails significantly skewed to one side. For example, most years / regions may have 0, 1, or 2 accidents; a few years / regions may have several major accidents (e.g., dozens); and a very few extreme events (e.g., mine disasters) may have hundreds. The long tail on the right side will significantly inflate the mean and increase the standard deviation, causing the normality assumption in the regression analysis to fail, and the estimation results may be dominated by a few extreme values.

[0028] Step 25: For the fixed effects model, it is necessary to verify that the core explanatory variables have sufficient time variation within individuals (i.e., the variables cannot be constant within individuals), and note that unbalanced panels can still be effectively estimated when the mean is removed within the group, but each individual is required to have at least two observation points.

[0029] Step 26 involves exploratory data analysis using descriptive statistics, correlation heatmaps, and visualization of within-group variation to ensure the data quality meets the estimation requirements of a two-way fixed effects model. The following exploratory data analysis verifies whether the data quality of the data processed in steps 21 to 25 meets the estimation requirements of a two-way fixed effects model: Verification of within-group sufficiency of variation: Plot line graphs of the core independent variables (historical meteorological data, urbanization rate data, total rural population) of each observed individual (such as administrative region) over time to confirm that the time series of each individual shows obvious fluctuations rather than an approximate horizontal straight line, and that the within-group standard deviation of each variable is greater than 0. Weak correlation verification: Calculate and visualize the Pearson correlation coefficient heatmap between independent variables. All pairs of independent variables must have an absolute value of correlation coefficient less than 0.8. If there is a strong correlation greater than 0.8, return to step 24 to reconsider variable selection or centering. Distribution pattern verification: By descriptive statistics outputting the skewness, kurtosis and quantile of each variable, it is confirmed that the absolute value of the skewness of the dependent variable (safety production accident data) and the continuous independent variable is less than 2 (less than 1 after logarithmic transformation), and there is no risk of extreme residuals exceeding the mean ± 3 times the standard deviation. If all three verifications above pass, the data quality is deemed to meet the estimation requirements of the two-way fixed effects model; otherwise, return to the corresponding preceding step (step 23, 24 or 25) for reprocessing based on the failed item.

[0030] Step 3: Classify the preprocessed data. This includes classifying the accident frequency from the safety production accident data. Num_ accident it CasualtiesCasualty it General accidents Minor_accident it Major accidents Moderate_ Accident it , serious accidents Major_accident it Defined as the dependent variable; the monthly average temperature in the historical meteorological data. Temperature it Monthly average temperature squared term Temperature it ² relative humidity Humidity it Cloud data Cloud it Rainfall data Rain it Snowfall data Snow it Fog concentration data Fog it Defined as explanatory variables; atmospheric pressure data from the historical meteorological data. Pressure it Wind speed data Windspeed it The product of temperature and humidity Humidity it × Temperature it Defined as a control variable.

[0031] Step 4: Use a two-way fixed effects model as the baseline model to analyze the impact of meteorology on the frequency and severity of rural work safety accidents.

[0032] Table 1 is created based on historical meteorological data and work safety accident data.

[0033] Table 1. The impact of weather on the number of workplace safety accidents and casualties at the grassroots level.

[0034] The values ​​in parentheses represent robust standard errors. p<0.01, p<0.05, p < 0.1. Analysis of the impact of historical meteorological data on workplace accident data in Table 1 reveals that the linear coefficient of temperature is significantly negative, while the quadratic coefficient is significantly positive, both indicating an inverted U-shaped relationship between temperature and the number of accidents: the number of accidents initially increases with rising temperature, but decreases after exceeding a critical value and continuing to rise. This is because extreme high or low temperatures trigger work stoppages or proactive safety interventions (such as increasing job rotation) to reduce risk. This suggests that moderate temperature increases significantly increase accident risk, and regulators should not only focus on extreme high temperatures but also pay more attention to the moderate temperature range, where risks are more prominent yet easily overlooked.

[0035] Furthermore, the humidity coefficient was significantly negative, indicating fewer accidents under high humidity. A plausible explanation is that increased humidity reduces static electricity buildup, enhances surface adhesion, and improves safety conditions. The drier the weather, the higher the frequency of accidents, suggesting that safety management and policy-making should place greater emphasis on dry weather. Similarly, cloud cover significantly reduces accidents by mitigating heat stress through sun protection and encouraging personnel to work cautiously on cloudy days.

[0036] The coefficients for rainfall, snowfall, and fog were all significantly positive. Rainfall mainly increases the risk by reducing visibility (<500 meters) and making the road surface slippery; snowfall exacerbates the danger due to reduced road friction and equipment freezing; fog reduces visibility to <100 meters, causing orientation obstacles in traffic operation spaces.

[0037] As shown in column (4) of Table 1, the air temperature, the square of the air temperature, and the humidity coefficient are all positive but not significant, indicating that changes in temperature and humidity have no significant impact on the total number of casualties. This may be because temperature and humidity mainly increase the probability of general accidents. The cloud cover coefficient is significantly negative, indicating fewer casualties on cloudy days. For every additional day of rainfall or fog, the number of casualties increases significantly by 193.1% and 374.6%, respectively.

[0038] To further examine the heterogeneity of meteorological impacts, the effects of meteorological variables on general accidents, major accidents, and serious accidents were analyzed. The results are shown in columns (1)-(3) of Table 2. The analysis shows that the effects of meteorological variables on accidents of different severity levels vary: temperature has an inverted U-shaped relationship with general accidents; humidity and cloud cover are significantly negatively correlated with general accidents; and rainfall, snowfall, and fog have significant positive effects on general accidents.

[0039] Table 2. The impact of weather on the severity of rural work safety accidents.

[0040] The values ​​in parentheses represent robust standard errors. p<0.01, p<0.05, p<0.1. Notably, only rainfall had a significant positive impact on major accidents; snowfall had a significant negative impact on serious and catastrophic accidents. This differential result reveals the dual effect of snowfall: snowfall increases operational difficulty and raises the risk of general accidents; however, it prompts preventative measures, slows down operations, and adjusts resources, thereby avoiding serious consequences and reducing the risk of serious and catastrophic accidents.

[0041] The above results indicate that meteorological conditions have a heterogeneous impact on accidents of varying severity. Therefore, when there are significant changes in temperature, humidity, and cloud cover, regulatory authorities should focus on preventing and controlling general accidents; during rainy weather, the focus should be on reducing the risk of larger accidents.

[0042] Based on the above analysis results, the baseline model is set as follows: ; in: Y it The dependent variables include accident frequency, number of casualties, number of general accidents, number of major accidents, and number of serious and major accidents; β 0 、β 1 、β 2 、β 3 、β 4 、β 5 、β 6 、β 7 、β 8 、β 9 and β 10 These are the weight coefficients for each variable; The core independent variable is the monthly meteorological variable of each province; Air pressure, wind speed, humidity × temperature are the controlled variables; λ i For provincial fixed effects, ν t For month-fixed effects, control for provincial characteristics that do not change over time and time-specific shocks; ε it This is a random disturbance term.

[0043] Step 5: Add the urbanization rate data and the total rural population as adjustment variables to the baseline model to establish a grassroots work safety accident risk assessment model.

[0044] Table 3 is created based on urbanization rate data and work safety accident data.

[0045] Table 3. Moderating effect of urbanization rate data on the number and severity of rural work safety accidents.

[0046] The values ​​in parentheses represent robust standard errors. p<0.01, p<0.05, p<0.1. Analysis of Table 3 shows that the temperature × urbanization interaction term in column (2) is significantly negative, indicating that the higher the urbanization rate, the weaker the impact of temperature on casualties. The humidity × urbanization interaction term in columns (1), (3), and (5) is significantly negative, indicating that higher urbanization rates can mitigate the impact of humidity on the number and severity of accidents. Conversely, the cloud cover × urbanization interaction term in column (2) is significantly positive, indicating that in areas with higher urbanization levels, cloudy days have a stronger impact on casualties. This pattern may reflect differences in exposure levels, infrastructure, and institutional capabilities at different urbanization levels. For example, high urbanization can provide better safety infrastructure and emergency response systems, reducing accidents related to high temperature and high humidity; however, dense urban populations and complex transportation networks can amplify the impact of cloudy days on casualties. Overall, urbanization can significantly alter the direction and intensity of meteorological impacts on occupational safety and health, suggesting that local urbanization background must be considered when formulating targeted prevention and control strategies.

[0047] Table 4 is created based on the total rural population and data on work safety accidents.

[0048] Table 4. Moderating effect of total rural population on the number and severity of rural work safety accidents.

[0049] The values ​​in parentheses represent robust standard errors. p<0.01, p<0.05, p < 0.1. Analysis of Table 4 shows that the temperature × population interaction term is significantly negative for the total number of accidents and general accidents, but significantly positive for major and catastrophic accidents. This indicates that regions with a higher population proportion experience fewer general accidents due to rising temperatures, but a higher probability of major and catastrophic accidents. This may be because rural laborers are more likely to engage in heavy outdoor physical labor and are more exposed to extreme heat, making accidents more likely to escalate into serious incidents. Furthermore, the cloud cover × population interaction term is significantly negative for the total number of accidents and general accidents, indicating that increased urbanization can mitigate the adverse effects of cloudy weather on general accidents. This may be because better urban infrastructure, lighting, and regulatory systems reduce the likelihood of general accidents occurring on cloudy days.

[0050] Based on the above analysis, a grassroots work safety accident risk assessment model is established, including an urbanization rate impact model and a rural population total impact model. The urbanization rate impact model is as follows: ; in, This refers to taking the urbanization rate data under t. the median of If the base is , Then the corresponding =1, otherwise 0; β 11 、β 12 and β 13 These are the weight coefficients for each variable; The model for the impact of the total rural population is as follows: ; in, This refers to the total rural population. β 14 、β 15 and β 16 These are the weighting coefficients for each variable.

[0051] Using these two models, we can explore how moderating variables (urbanization rate data, rural population) enhance or weaken the impact of weather on dependent variables (number of work safety accidents, number of casualties, severity), and dynamically formulate early warning and prevention recommendations based on the specific laws of the moderating effect.

[0052] Accordingly, embodiments of the present invention also provide a grassroots work safety accident risk assessment system oriented towards meteorological disaster-causing factors, such as... Figure 2 The diagram shown is a structural schematic of the system. This grassroots safety production accident risk assessment system for meteorological disaster-causing factors includes the following modules: Data acquisition unit 201 collects historical meteorological data, work safety accident data, urbanization rate data, and total rural population. The data preprocessing unit 202 preprocesses the collected historical meteorological data, work safety accident data, urbanization rate data, and total rural population. Data classification unit 203 classifies the preprocessed historical meteorological data and the work safety accident data; Benchmark model construction unit 204 uses a two-way fixed effects model as the benchmark model to analyze the impact of meteorology on the frequency and severity of grassroots work safety accidents; The accident risk assessment model construction unit 205 incorporates the urbanization rate data and the total rural population as adjustment variables into the baseline model to establish a grassroots work safety accident risk assessment model.

[0053] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0054] The present invention also provides a storage medium, which is a computer-readable storage medium storing a computer program thereon, the computer program being executable when it runs. Figure 1 The method shown may include some or all of the steps. The storage medium may include read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, etc. The storage medium may also include non-volatile memory or non-transitory memory, etc.

[0055] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data provider to another website, computer, server, or data provider via wired or wireless means.

[0056] The embodiments of the present invention have been described in detail above. Specific implementation methods have been used to illustrate the present invention. The descriptions of the embodiments above are only for the purpose of helping to understand the methods and systems of the present invention, and are merely some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention, and the content of this specification should not be construed as a limitation of the present invention. Therefore, any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for risk assessment of grassroots work safety accidents oriented towards meteorological disaster-causing factors, characterized in that, The method includes: Step 1: Collect historical meteorological data, work safety accident data, urbanization rate data, and total rural population. Step 2: Perform data preprocessing on the collected historical meteorological data, work safety accident data, urbanization rate data, and total rural population. Step 3: Classify the preprocessed historical meteorological data and the work safety accident data; Step 4: Use a two-way fixed effects model as the baseline model to analyze the impact of meteorology on the frequency and severity of grassroots work safety accidents; Step 5: Add the urbanization rate data and the total rural population as adjustment variables to the baseline model to establish a grassroots work safety accident risk assessment model.

2. The method for risk assessment of grassroots work safety accidents based on meteorological disaster-causing factors as described in claim 1, characterized in that, In step 1, the historical meteorological data includes temperature data, humidity data, atmospheric pressure data, wind speed data, cloud cover data, rainfall data, snowfall data, and fog concentration data; the work safety accident data includes the accident time, accident location, accident frequency, accident cause, number of deaths, and number of injuries; accidents are classified according to their causes; accidents are divided into three levels according to the number of deaths and injuries: accidents with fewer than 3 deaths or fewer than 10 injuries are classified as general accidents; accidents causing 3 to 10 deaths or 10 to 50 serious injuries are classified as major accidents; accidents causing more than 10 deaths or more than 50 injuries are classified as serious and extremely serious accidents; the accident frequency is measured by the number of accidents per month.

3. The method for risk assessment of grassroots work safety accidents based on meteorological disaster-causing factors as described in claim 2, characterized in that, In step 1, the historical meteorological data is collected on a monthly basis, including the monthly average temperature. Temperature it and the square of the monthly average temperature Temperature it ² Cloud data Cloud it The monthly average total cloud cover is calculated as the percentage of the sky obscured by clouds, ranging from 0% (clear sky) to 100% (overcast); rainfall data. Rain it Includes light rain, moderate rain, heavy rain, showers, thunderstorms, and the number of rainy days; snowfall data. Snow it Includes data on light snow, snow, heavy snow, sleet, ice pellets, and the number of days with snowfall; fog concentration data. Fog it Includes light fog, dense fog, heavy fog, and the number of days with fogfall; atmospheric pressure data. Pressure it The monthly average sea-level pressure, quantified in hectopascals, reflects the monthly atmospheric pressure changes; wind speed data... Windspeed it Monthly average wind speed, measured in meters per second, represents the average airflow velocity at the study location; humidity data includes relative humidity. Humidity it The product of temperature and humidity Humidity it ×Temperature it The humidity-temperature product term is calculated by multiplying the relative humidity by the temperature. Humidity it In percentage terms.

4. The method for risk assessment of grassroots work safety accidents based on meteorological disaster-causing factors as described in claim 2, characterized in that, In step 1, the urbanization rate data and total population are collected, and the total rural population is calculated: Total rural population = Total population (1 - Urbanization rate data).

5. The method for risk assessment of grassroots work safety accidents based on meteorological disaster-causing factors according to claim 1, characterized in that, Step 2 includes: Step 21: Clean the collected historical meteorological data, work safety accident data, urbanization rate data, and total rural population data by standardizing date and character formats, removing duplicate observations, standardizing variable naming, and specifying individual identifiers and time identifiers. Step 22: When the historical meteorological data, the work safety accident data, the urbanization rate data, and the total rural population have missing values ​​after data cleaning, if the missing value is a key variable, the observation row containing the missing value is directly deleted. Step 23: Perform data detection and outlier processing on the historical meteorological data, the work safety accident data, the urbanization rate data, and the total rural population after missing value processing. Use box plots or interquartile ranges to identify extreme values, remove obviously erroneous values, and perform tail reduction or logarithmic transformation on real extreme values ​​to reduce interference with the mean-free estimation within the group. Step 24: Perform variable transformation on the historical meteorological data, work safety accident data, urbanization rate data, and total rural population after outlier processing. If the variable distribution is severely skewed, logarithmic transformation is used. At the same time, check for multicollinearity between interaction terms and original variables, calculate the variance inflation factor (VIF), and if the VIF exceeds 10, perform variable centering or delete redundant terms. Step 25: Verify whether the historical meteorological data has sufficient temporal variation within an individual, i.e., each individual must have at least two observation points at time. If the condition is not met, delete the data. Step 26: Exploratory data analysis is conducted through descriptive statistics, correlation heatmaps, and visualization of within-group variation to ensure that the data quality of the historical meteorological data, the work safety accident data, the urbanization rate data, and the total rural population meets the estimation requirements of the two-way fixed effects model.

6. The method for risk assessment of grassroots work safety accidents based on meteorological disaster-causing factors as described in claim 3, characterized in that, In step 3, the accident frequency, number of casualties, number of general accidents, number of major accidents, and number of serious and major accidents in the work safety accident data are defined as the dependent variables; the monthly average temperature in the historical meteorological data is defined as the dependent variable. Temperature it Monthly average temperature squared term Temperature it ² relative humidity Humidity it Cloud data Cloud it Rainfall data Rain it Snowfall data Snow it Fog concentration data Fog it Defined as explanatory variables; atmospheric pressure data from the historical meteorological data. Pressure it 、 Wind speed data Windspeed it The product of temperature and humidity Humidity it × Temperature it Defined as a control variable.

7. The method for risk assessment of grassroots work safety accidents based on meteorological disaster-causing factors as described in claim 6, characterized in that, In step 4, the baseline model is set as follows: ； In the formula, Y it The dependent variables include accident frequency, number of casualties, number of general accidents, number of major accidents, and number of serious and major accidents; β 0 、β 1 、β 2 、β 3 、β 4 、β 5 、β 6 、β 7 、β 8 、β 9 and β 10 These are the weight coefficients for each variable; λ i For regional fixed effects, ν t For month-fixed effects, control for time-invariant regional characteristics and time-specific shocks; ε it This is a random disturbance term.

8. The method for risk assessment of grassroots work safety accidents based on meteorological disaster-causing factors as described in claim 7, characterized in that, In step 5, the grassroots work safety accident risk assessment model includes an urbanization rate impact model and a rural population total impact model. The urbanization rate impact model is as follows: ； in, This refers to taking t Urbanization rate data the median of If the base is , Then the corresponding =1, otherwise 0; β 11 、β 12 and β 13 These are the weight coefficients for each variable; The model for the impact of the total rural population is as follows: ； in, This refers to the total rural population. β 14 、β 15 and β 16 These are the weighting coefficients for each variable.

9. A grassroots work safety accident risk assessment system oriented towards meteorological disaster-causing factors, characterized in that, The system includes: The data acquisition unit collects historical meteorological data, work safety accident data, urbanization rate data, and total rural population. The data preprocessing unit preprocesses the collected historical meteorological data, work safety accident data, urbanization rate data, and total rural population. The data classification unit classifies the preprocessed historical meteorological data and the work safety accident data. The baseline model construction unit uses a two-way fixed effects model as the baseline model to analyze the impact of meteorology on the frequency and severity of grassroots work safety accidents; The accident risk assessment model construction unit incorporates the urbanization rate data and the total rural population as adjustment variables into the baseline model to establish a grassroots work safety accident risk assessment model.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is run by the processor, it performs the steps of the grassroots work safety accident risk assessment method according to any one of claims 1 to 8, which is oriented towards meteorological disaster-causing factors.

Images