Transportation enterprise risk prediction method and system based on dangerous chemical risk image

Abstract

The application provides a transportation enterprise risk prediction method and system based on a dangerous chemical risk portrait, which comprises the following steps: obtaining enterprise transportation data from a road transportation credit information database; constructing a transportation enterprise risk evaluation index set; calculating the weight of each evaluation index in the enterprise risk evaluation index set according to the obtained enterprise transportation data; combining the entropy weight of each index in the evaluation index weight to calculate the positive and negative ideal solution distance and the closeness of the evaluation index, and obtaining the transportation risk ranking of the enterprise in each month through sorting to complete the construction of the dangerous chemical risk portrait and the risk prediction of the transportation enterprise based on the dangerous chemical risk portrait.

Description

Technical Field

[0001] This invention belongs to the field of transportation data prediction technology, and in particular relates to a method and system for predicting the risks of transportation enterprises based on the risk profile of hazardous chemicals. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] Related studies indicate that road transportation is the link in the chain with the highest risk of hazardous chemical leaks. Furthermore, because hazardous chemical transport often traverses densely populated areas, leaks during transport can easily trigger secondary accidents such as fires, explosions, and poisoning, causing serious harm to nearby residents, as shown in Table 1. With the rapid development of the chemical industry, the accident rate of hazardous chemicals remains high and shows a year-on-year upward trend. The frequent occurrence of these serious accidents severely damages the social image of hazardous chemical transport companies, threatens the safety of people's lives and property, and has attracted high attention from society and the media.

[0004] Table 1 Typical Hazardous Chemical Transportation Accidents

[0005]

[0006]

[0007] Accident statistics show that human error is the primary cause of hazardous chemical transport accidents. Furthermore, some transport companies have inadequate safety management, fail to fulfill their corporate responsibilities, and have numerous management loopholes, further exacerbating the occurrence of hazardous chemical transport accidents.

[0008] To create an efficient regulatory mechanism for hazardous chemical transport companies, some regions have introduced "pass codes" that monitor vehicle driving trajectories and violation records throughout the entire journey, and dynamically score drivers for dangerous driving behaviors. Based on the total score throughout the driving process, a three-color "red, yellow, and green" pass code is generated to intuitively display the driver's safety risk level. However, the dynamic scoring rules of this regulatory mechanism are based on the subjective judgment of relevant management personnel, and the determination of the weights of various evaluation indicators and the reliability of the scoring mechanism need further verification.

[0009] Furthermore, existing research on corporate risk mostly focuses on corporate financial credit risk, and there is a lack of research on risk assessment based on the characteristics of the enterprise itself to construct a set of characteristic indicators.

[0010] Accurate evaluation of corporate customer credit requires scientific and reasonable analytical methods and models, and scholars both domestically and internationally have achieved considerable research results. Since the 1960s, European and American countries have begun to study corporate credit rating methods, establishing univariate discriminant models, multivariate discriminant analysis models, and logistic regression models for corporate credit rating and financial crisis early warning research.

[0011] Multivariate discriminant analysis (MDA) has found some application in enterprise risk assessment due to its simple principle and excellent performance. Zou Xinyue et al. used typical multivariate linear discriminant analysis to study the credit risk of 128 listed companies in my country, showing that the method has strong predictive ability. Maryam et al. used MDA to predict bankruptcy for all manufacturing companies listed on the Tehran Stock Exchange from 2000 to 2010, confirming that the method has high accuracy in identifying bankrupt companies. However, MDA models require strict assumptions such as normal distribution, which limits the applicability of the model to some extent. Therefore, Berkson et al. proposed a Logistic regression model based on MDA. Subsequently, Engelmann et al., based on a large sample of SME data, empirically compared the predictive performance of the improved MDA model and the Logistic regression model, showing that the Logistic regression model has significant advantages.

[0012] Logistic regression models are highly adaptable and do not have strict limitations regarding whether data follows a normal distribution, but their computational accuracy is lower than that of emerging artificial intelligence models. Common artificial intelligence models include decision trees (DT), case-based reasoning (CBR), artificial neural networks (ANN), and support vector machines (SVM). Guan Qihai et al., based on a massive database of loan enterprises across Chinese financial institutions, constructed and empirically explored default discrimination models for short-term loan enterprises in the manufacturing industry, categorized by size and region. They compared and analyzed the accuracy of multivariate discriminant analysis models, logistic models, and neural network models, and presented the optimal default discrimination model for manufacturing enterprises categorized by size and region. The results show that artificial intelligence models have higher accuracy than other models, but the former suffers from poor stability and poor business interpretability, making it difficult for researchers to grasp the internal logic of its credit assessment.

[0013] In recent years, TOPSIS, as a widely accepted multi-attribute decision-making technique, has been gradually applied to various fields of risk assessment. Kustiyahningsihden et al., based on the determination of indicator weights using the Fuzzy Analytic Hierarchy Process (FHAP) and combined with the TOPSIS method, ranked the risks of small and medium-sized enterprises (SMEs) and proposed improvement suggestions for the operation and management of SMEs in the context of the global pandemic. Cheng et al. used the Fermatean fuzzy TOPSIS method to measure the relative closeness of 10 Chinese quantum communication listed companies to the ideal solution, providing credit risk assessments for each company. Dong et al. first established a comprehensive evaluation index system for thermal power generating units with 20 secondary indicators from four aspects: reliability indicators, economic indicators, technical supervision indicators, and main operating indicators. They then combined the entropy weight-TOPSIS method with grey relational analysis to effectively evaluate the operation of thermal power generating units.

[0014] A review of existing literature reveals a lack of research on the risks of freight companies, primarily focusing on financial risk assessment rather than considering the specific characteristics of their transportation operations. Furthermore, previous studies have largely employed static credit ratings, emphasizing only the initial rating of the evaluated entity and failing to provide dynamic credit ranking.

[0015] In summary, current enterprise risk assessments do not make reasonable use of information technology, are highly subjective in the assessment process, lack objective models for prediction, and have low prediction accuracy. Summary of the Invention

[0016] To overcome the shortcomings of the existing technology, this invention provides a risk prediction method for transportation companies based on the risk profile of hazardous chemicals, establishes a reasonable and comprehensive risk assessment index set, constructs a reliable and effective risk assessment model, and provides relatively accurate predictions.

[0017] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions:

[0018] Firstly, a risk prediction method for transportation companies based on hazardous chemical risk profiles was disclosed, including:

[0019] Obtain enterprise transportation data from the road transport credit information database;

[0020] Construct a risk assessment indicator set for transportation enterprises;

[0021] The weights of each evaluation indicator in the risk assessment indicator set for the enterprise are calculated based on the acquired enterprise transportation data.

[0022] By combining the entropy weight of each indicator in the evaluation index weight, the distance and proximity of the positive and negative ideal solutions of the evaluation index are calculated. After sorting, the transportation risk ranking of the enterprise in each month is obtained, and the risk profile of hazardous chemicals is constructed. Based on the risk profile of hazardous chemicals, the risk of transportation enterprises is predicted.

[0023] As a further technical solution, a risk assessment index set for transportation enterprises is constructed by selecting fatigue warnings, distraction warnings, speeding, poor driving, mileage, forward collision warnings, offline warnings, pedestrian collision warnings, monthly waybill counts, number of transport vehicles, and hazardous chemical categories from the dimensions of driving performance, vehicle driving conditions, and hazardous goods.

[0024] As a further technical solution, the steps of quantifying the risks of dangerous goods are also included:

[0025] The risk coefficient R is the sum of the ratios of the actual quantity of each hazardous chemical in the unit (i.e., the online quantity) to the critical quantity specified in "Identification of Major Hazard Installations of Hazardous Chemicals" (GB18218), after correction by a correction factor.

[0026] As a further technical solution, when calculating the weights of each evaluation indicator in the enterprise's risk assessment indicator set, the entropy weight method is used to calculate the weights of the transportation enterprise's risk indicators. The specific calculation steps are as follows:

[0027] For n samples and m indicators, then x ij Let be the value of the j-th indicator for the i-th sample;

[0028] Normalization of indicators: homogenization of heterogeneous indicators;

[0029] Calculate the proportion of the i-th sample value under the j-th indicator to that indicator;

[0030] Calculate the entropy value of the j-th index;

[0031] Calculate information entropy redundancy;

[0032] Calculate the weight of each indicator.

[0033] As a further technical solution, the distance and proximity between the positive and negative ideal solutions of the evaluation index are calculated. The specific process is as follows:

[0034] Construct a standardized decision matrix;

[0035] Construct a weighted standardized decision matrix;

[0036] Determine the positive and negative ideal solutions;

[0037] Calculate the distance from each alternative solution to the positive and negative ideal solutions;

[0038] Calculate the scheme C that is closest to the ideal solution. h ;

[0039] Sorting the options according to C h Arranged in descending order.

[0040] Secondly, a risk prediction system for transportation companies based on hazardous chemical risk profiles was disclosed, including:

[0041] The data acquisition module is configured to: acquire enterprise transportation data from the road transport credit information database;

[0042] The indicator set construction module is configured to: construct a risk assessment indicator set for transportation enterprises;

[0043] The evaluation indicator weight calculation module is configured to: calculate the weights of each evaluation indicator in the risk evaluation indicator set for the acquired enterprise transportation data;

[0044] The risk prediction module is configured to: calculate the distance and proximity of the positive and negative ideal solutions of the evaluation indicators by combining the entropy weight of each indicator in the evaluation index weight, obtain the transportation risk ranking of the enterprise in each month after sorting, complete the construction of the hazardous chemical risk profile, and make risk prediction for the transportation enterprise based on the hazardous chemical risk profile.

[0045] The above one or more technical solutions have the following beneficial effects:

[0046] For companies engaged in the transportation of hazardous chemicals, their management practices directly impact the behavior of their drivers and vehicles, thus influencing the occurrence of hazardous chemical transportation accidents. Considering the unique characteristics of these companies, this invention establishes a reasonable and comprehensive set of risk assessment indicators, constructs a reliable and effective risk assessment model, and provides timely predictive information to company managers and transportation regulatory authorities. This is of great significance for both company management and road safety.

[0047] Based on the characteristics of enterprises, this invention constructs a risk assessment index set for hazardous chemical transportation enterprises from three aspects: driving performance, vehicle driving conditions, and hazardous chemicals. It builds an entropy weight-TOPSIS model, uses the entropy weight method to determine the weight of each index, and uses the TOPSIS model to complete the risk score ranking of transportation enterprises. Furthermore, it explores the relationship between various factors and enterprise risk, which can accurately predict the transportation risk information of enterprises.

[0048] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0049] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0050] Figure 1 This is a risk trend chart for a company in each month, as shown in an embodiment of the present invention. Detailed Implementation

[0051] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0052] It should be noted that the terminology used herein is for the purpose of describing particular implementations only and is not intended to limit the exemplary implementations of the present invention.

[0053] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0054] Example 1

[0055] This embodiment discloses a risk prediction method for transportation companies based on hazardous chemical risk profiles, including:

[0056] Obtain enterprise transportation data from the road transport credit information database;

[0057] Construct a risk assessment indicator set for transportation enterprises;

[0058] The weights of each evaluation indicator in the risk assessment indicator set for the enterprise are calculated based on the acquired enterprise transportation data.

[0059] By combining the entropy weight of each indicator in the evaluation index weight, the distance and proximity of the positive and negative ideal solutions of the evaluation index are calculated. After sorting, the transportation risk ranking of the enterprise in each month is obtained, and the risk profile of hazardous chemicals is constructed. Based on the risk profile of hazardous chemicals, the risk of transportation enterprises is predicted.

[0060] This invention effectively reduces the accident rate of hazardous chemical transport vehicles and improves the regulatory efficiency of hazardous chemical transport companies through the above-mentioned technical solution.

[0061] Regarding the selection of evaluation indicators:

[0062] The risks of hazardous chemical transportation companies are influenced by many factors, especially company-specific indicators such as driver performance and vehicle transport conditions. Considering the comprehensiveness, importance, scientific nature, and operability of the selected indicators, the technical solution of this invention selects 11 sub-items from three dimensions—driving performance, vehicle operating conditions, and hazardous goods—to construct a risk assessment indicator set for transportation companies, as shown in Table 2.

[0063] Table 2 Set of Risk Quantification Indicators for Transportation Enterprises

[0064]

[0065]

[0066] Risk quantification of dangerous goods:

[0067] For dangerous goods transported by operating companies, databases generally only contain the corresponding names and classifications in international dangerous goods classifications, without specific numerical quantifications of the risks inherent in the dangerous goods themselves. Therefore, the technical solution of this invention uses the sum of the ratios R between the actual quantity (i.e., online quantity) of various dangerous chemicals within a unit and the critical quantity specified in "Identification of Major Hazard Installations of Hazardous Chemicals" (GB18218) and corrected by a correction factor as the risk coefficient value for the dangerous chemicals.

[0068]

[0069] In the formula:

[0070] q1,q2,...,q n —Actual quantity (online) of each hazardous chemical (unit: tons);

[0071] Q1,Q2,...,Q n —Critical quantities (in tons) corresponding to each hazardous chemical;

[0072] β1,β2,...,β n —Correction factors corresponding to each hazardous chemical;

[0073] α—Correction factor for personnel exposed to hazardous chemicals in the plant area of ​​a major hazard source.

[0074] The correction factor β values ​​for different categories of hazardous chemicals are given in Tables 3 and 4. Furthermore, since the technical solution of this invention does not involve personnel exposed outside the factory, α = 1.

[0075] Table 3. Values ​​of Correction Coefficient β

[0076]

[0077] Note: The categories of hazardous chemicals are based on the "List of Dangerous Goods".

[27] Classification criteria determined

[0078] Table 4. Values ​​of Correction Factor β for Common Toxic Gases

[0079]

[0080] Note: For toxic gases not listed in Table 4, a value of β = 2 can be used; for highly toxic gases, a value of β = 4 can be used.

[0081] Determining the weights of evaluation indicators:

[0082] Methods for calculating indicator weights include entropy weighting, analytic hierarchy process (AHP), and principal component analysis (PCA). Among these, entropy weighting utilizes the decision matrix and the output entropy of each indicator to determine its weight coefficient. The weight coefficient measures the relative intensity of competition among the evaluation indicators, rather than their importance, thus offering higher reliability and accuracy compared to subjective weighting. This invention employs entropy weighting to calculate the risk indicator weights for transportation enterprises. The specific calculation steps are as follows:

[0083] ① For n samples and m indicators, then x ij Let be the value of the j-th indicator for the i-th sample;

[0084] ② Normalization of indicators: Homogenization of heterogeneous indicators

[0085] Positive indicators:

[0086]

[0087] Negative indicators:

[0088]

[0089] Where, x ij Let be the value of the j-th indicator for the i-th sample;

[0090] ③ Calculate the proportion of the i-th sample value under the j-th indicator to that indicator.

[0091]

[0092] Where n is the number of enterprise samples and m is the number of indicators;

[0093] ④ Calculate the entropy value of the j-th index.

[0094]

[0095] Where k = 1 / ln(n) > 0, satisfying e j ≥0;

[0096] ⑤ Calculate the information entropy redundancy (difference).

[0097] d j =1-e j ,j=1,...,m (6)

[0098] ⑥ Calculate the weight of each indicator

[0099]

[0100] Where, x ij This is the standardized data.

[0101] TOPSIS-based risk assessment model for transportation companies:

[0102] TOPSIS is a ranking method known as approximation of the ideal solution. It is often used to solve multi-attribute decision problems. Its principle is to use distance metrics to measure the closeness of evaluation objects to an idealized target, thereby ranking and selecting the appropriate solution. The basic idea of ​​this method is to assume positive and negative ideal solutions, calculate the distance between each sample and these ideal solutions, obtain its relative closeness to the ideal solution, and then rank the evaluation objects accordingly. The specific steps are described below:

[0103] ① Construct a standardized decision matrix

[0104]

[0105] Where, r hj Let j represent the index value of the h-th option.

[0106] ② Construct a weighted standardized decision matrix

[0107] V = [w j r hj ] = [v hj ] z×m h = 1, 2, ..., zj = 1, 2, ..., n (4)

[0108] Among them, w j It is the weight of index j determined by the entropy weight method.

[0109] ③ Determine the positive and negative ideal solutions

[0110]

[0111] Among them, A + Represents the ideal solution, A - This represents the negative ideal solution.

[0112] ④ Calculate the distance from each alternative solution to the positive and negative ideal solutions.

[0113]

[0114] in, This represents the distance from solution h to the ideal solution. Let represent the distance from solution h to the negative ideal solution.

[0115] ⑤ Calculate the scheme that is closest to the ideal solution.

[0116]

[0117] ⑥ The schemes are ranked according to C. h Arranged in descending order.

[0118] Case Analysis

[0119] This example uses a hazardous chemical transportation company in a province in East China for verification. Established in 2018, the company has 271 drivers and 133 vehicles for transporting hazardous chemicals, primarily handling Class 2.1, Class 2.3, Class 3, Class 5, Class 6, and Class 8 hazardous chemicals. The empirical data is taken from the provincial road transport credit information database, mainly focusing on data from July 2019 to September 2021. Table 5 presents descriptive statistics of the company's relevant evaluation indicators.

[0120] Table 5. Descriptive Statistics of Enterprises

[0121]

[0122] The weights of each evaluation indicator for the company were calculated using the entropy weight method, and the results are shown in Table 6. It can be seen that the entropy weight of mileage is 31.89%, ranking first among all indicators. This indicates that this indicator has the greatest impact on the company's risk assessment and is a factor that needs to be considered when transportation companies conduct risk assessments. Secondly, the weights of fatigue warnings, forward collisions, and distraction incidents are also relatively large, indicating that driver performance also has a significant impact on the company's operational risk. This is consistent with the conclusions of existing research on the causal analysis of hazardous chemical transportation accidents.

[0123] Table 6. Results of Indicator Weights

[0124]

[0125] Based on formulas (5), (6), and (7) and combined with the entropy weight of each indicator, the distance and proximity of the positive and negative ideal solutions of the evaluation indicators are calculated. After sorting, the transportation risk ranking of the enterprise in each month is obtained, as shown in Table 7. Among them, the months with the highest and lowest transportation risk for the enterprise are August 2020 and February 2020, respectively. Figure 1 The report presents the company's monthly mileage ranking and risk assessment results. It can be seen that the company's overall risk ranking and monthly mileage ranking show a similar trend. Therefore, for this company, reducing operational risks should begin with addressing vehicle driving conditions, including timely vehicle inspection and maintenance to ensure safe driving.

[0126] Table 7 Risk Assessment Results of Transportation Enterprises

[0127]

[0128]

[0129] from Figure 1 The average rankings for each quarter show that transportation companies face the lowest transportation risks in the first quarter of each year. This may be due to the impact of the Spring Festival holiday. However, the risks faced by these companies in the first quarter of 2020 were significantly lower than in the first quarter of 2021, possibly related to my country's epidemic prevention and control policies in 2020. During the global pandemic, the control measures on the logistics industry limited the volume of goods transported, and the risks for companies increased significantly in the second and third quarters. This may be because the lifting of lockdown policies greatly stimulated the development of the logistics industry, leading to a surge in order volume and vehicle mileage.

[0130] Statistical analysis of enterprise risks on a quarterly basis reveals that enterprises are likely to face lower business volume and vehicle mileage in the first quarter of the year due to holidays, thus facing less enterprise risk compared to other quarters. Future research could further consider the risk situation of enterprises during holidays and non-holiday periods. During holidays, changes in toll collection policies and control measures may lead to different arrangements for vehicle routes and travel times by transportation companies, requiring further detailed research.

[0131] This invention focuses on hazardous materials transportation companies. From the perspective of their characteristics, it establishes a risk assessment index set based on three dimensions: driving performance, vehicle operating conditions, and hazardous materials. The risk of hazardous materials is quantified, and then the entropy weight-TOPSIS method is used to construct a risk profile of the hazardous materials transportation company, thus achieving risk assessment. The conclusions of this invention are as follows:

[0132] (1) Analysis of the information content of each indicator using the entropy weight method revealed that the indicator with the greatest impact on transportation risk is the vehicle's mileage, with a weight of 31.89%. The weights of the indicators of fatigue warnings, forward collisions, and distraction in driving performance are also greater than 10%. For transportation company managers, it is advisable to strengthen daily training for drivers, improve the driving safety awareness of internal employees, pay attention to vehicle maintenance, and promptly check the mechanical condition of vehicles to avoid accidents.

[0133] (2) The TOPSIS method analysis showed that the company had the lowest transportation risk in February 2020, with the lowest number of fatigue warnings, pedestrian collisions, and mileage compared to other months; the transportation risk was the highest in August of the same year, with the highest number of bad driving behaviors, offline warnings, and mileage, which is consistent with the conclusion on the importance of indicators obtained by the entropy weight method.

[0134] (3) Analysis of the risk assessment results of enterprises from 2019 to 2021 revealed that the risk was low in the third and fourth quarters of 2019, which was largely due to the fact that the enterprises were in the initial stage at that time. In 2020, except for the first quarter, the transportation risk in other quarters was at a high level, which may be due to the influence of relevant policies in the early 2020. The transportation risk of enterprises increased significantly in the third and fourth quarters, which is likely because the relevant policies in the early stage stimulated the transportation volume in the later stage.

[0135] When conducting risk assessments for transportation companies, focusing solely on credit and financial risks is far from sufficient and cannot eliminate the adverse consequences and negative public opinion arising from traffic accidents involving internal vehicles of hazardous chemical transportation companies. To provide guidance and recommendations for the overall operation of transportation companies, it is essential to start from the perspective of the company's own operational characteristics, accurately depict its risk profile, obtain accurate predictive information, and provide corresponding operational management suggestions based on the identified risk sources.

[0136] Based on the characteristics of hazardous chemical transportation companies, the technical solution of this invention selects 11 indicators from three aspects: driving performance, vehicle driving status, and hazardous chemical goods to construct a risk assessment system for the company. The entropy weight-TOPSIS method is used to determine the weight of each indicator to realize the risk profile of the transportation company.

[0137] Example 2

[0138] The purpose of this embodiment is to provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the above-described method.

[0139] Example 3

[0140] The purpose of this embodiment is to provide a computer-readable storage medium.

[0141] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the steps of the above method.

[0142] Example 4

[0143] The purpose of this embodiment is to provide a risk prediction system for transportation companies based on hazardous chemical risk profiles, including:

[0144] The data acquisition module is configured to: acquire enterprise transportation data from the road transport credit information database;

[0145] The indicator set construction module is configured to: construct a risk assessment indicator set for transportation enterprises;

[0146] The evaluation indicator weight calculation module is configured to: calculate the weights of each evaluation indicator in the risk evaluation indicator set for the acquired enterprise transportation data;

[0147] The risk prediction module is configured to: calculate the distance and proximity of the positive and negative ideal solutions of the evaluation indicators by combining the entropy weight of each indicator in the evaluation index weight, obtain the transportation risk ranking of the enterprise in each month after sorting, complete the construction of the hazardous chemical risk profile, and make risk prediction for the transportation enterprise based on the hazardous chemical risk profile.

[0148] In a specific implementation example, the indicator set construction module selects fatigue warnings, distraction warnings, speeding, bad driving, mileage, forward collision warnings, offline warnings, pedestrian collision warnings, monthly waybills, number of transport vehicles, and hazardous chemical categories from the dimensions of driving performance, vehicle driving conditions, and hazardous goods to construct a risk assessment indicator set for transportation enterprises.

[0149] In a specific implementation example, a risk quantification module for dangerous goods is also included, which is configured as follows:

[0150] The risk coefficient R is the sum of the ratios of the actual quantity of each hazardous chemical in the unit (i.e., the online quantity) to the critical quantity specified in "Identification of Major Hazard Installations of Hazardous Chemicals" (GB18218), after correction by a correction factor.

[0151] In a specific implementation example, when the aforementioned evaluation index weight calculation module is used to calculate the weights of each evaluation index in the enterprise's risk assessment index set, the entropy weight method is selected to calculate the risk index weights of the transportation enterprise. The specific calculation steps are as follows:

[0152] For n samples and m indicators, then x ij Let be the value of the j-th indicator for the i-th sample;

[0153] Normalization of indicators: homogenization of heterogeneous indicators;

[0154] Calculate the proportion of the i-th sample value under the j-th indicator to that indicator;

[0155] Calculate the entropy value of the j-th index;

[0156] Calculate information entropy redundancy;

[0157] Calculate the weight of each indicator.

[0158] The steps and methods involved in the apparatuses of Embodiments 2, 3, and 4 above correspond to those in Embodiment 1. For specific implementation details, please refer to the relevant description section of Embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood as including any medium capable of storing, encoding, or carrying an instruction set for execution by a processor and enabling the processor to perform any of the methods in this invention.

[0159] Those skilled in the art will understand that the modules or steps of the present invention described above can be implemented using general-purpose computer devices. Optionally, they can be implemented using computer-executable program code, thereby allowing them to be stored in a storage device for execution by a computer device, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. The present invention is not limited to any particular combination of hardware and software.

[0160] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A risk prediction method for transportation enterprises based on hazardous chemical risk profiling, characterized by: include: The transportation data of enterprises is obtained from the road transport credit information database, including driving performance, vehicle driving status, and dangerous goods. Construct a risk assessment index set for transportation enterprises; select the following indicators from the dimensions of driving performance, vehicle driving conditions, and hazardous goods: number of fatigue warnings, number of distractions, number of speeding, number of poor driving, mileage, number of forward collisions, number of offline warnings, number of pedestrian collision warnings, number of monthly waybills, number of transport vehicles, and category of hazardous chemicals to construct the risk assessment index set for transportation enterprises; It also includes steps for quantifying the risks of dangerous goods: The sum of ratios calculated using the actual quantity (online quantity) of various hazardous chemicals within a unit and the critical quantity specified in the identification of major hazard sources of hazardous chemicals, after correction by a correction factor. As a risk coefficient value for hazardous chemicals; The risk profile of hazardous chemical transportation companies is constructed using the entropy weight-TOPSIS method to achieve risk assessment of transportation companies; The weights of each evaluation indicator in the enterprise's risk assessment indicator set are calculated based on the acquired enterprise transportation data. When calculating the weights of each evaluation indicator in the enterprise's risk assessment indicator set, the entropy weight method is used to calculate the weights of the transportation enterprise's risk indicators. The specific calculation steps are as follows: right One sample, Each indicator, then For the first The first sample The value of each indicator; Normalization of indicators: homogenization of heterogeneous indicators; Calculate the first The first item under the indicator The proportion of each sample value to the indicator; Calculate the first The entropy value of the indicator; Calculate information entropy redundancy; Calculate the weights of each indicator; By combining the entropy weight of each indicator in the evaluation index weight, the distance and proximity of the positive and negative ideal solutions of the evaluation index are calculated, and the merits and demerits of each evaluation index are ranked. After ranking, the transportation risk ranking of the enterprise in each month is obtained, and the risk profile of hazardous chemicals is constructed. Based on the risk profile of hazardous chemicals, the risk of transportation enterprises is predicted. The calculation process for the positive and negative ideal solutions of the evaluation index, as well as their closeness, is as follows: Construct a standardized decision matrix; Construct a weighted standardized decision matrix; Determine the positive and negative ideal solutions; Calculate the distance from each alternative solution to the positive and negative ideal solutions; Calculate the scheme that is closest to the ideal solution ; Scheme sorting, according to Arranged in descending order.

2. A risk prediction system for transportation enterprises based on hazardous chemical risk profiles, which executes the steps of the method described in claim 1 above, characterized in that: include: The data acquisition module is configured to: acquire enterprise transportation data from the road transport credit information database; The indicator set construction module is configured to: construct a risk assessment indicator set for transportation enterprises; The evaluation indicator weight calculation module is configured to calculate the weights of each evaluation indicator in the risk evaluation indicator set for the acquired enterprise transportation data. The risk prediction module is configured to: calculate the distance and proximity of the positive and negative ideal solutions of the evaluation indicators by combining the entropy weight of each indicator in the evaluation index weight, obtain the transportation risk ranking of the enterprise in each month after sorting, complete the construction of the hazardous chemical risk profile, and make risk prediction for the transportation enterprise based on the hazardous chemical risk profile.

3. The risk prediction system for transportation enterprises based on hazardous chemical risk profiles as described in claim 2, characterized in that, The indicator set construction module selects fatigue warnings, distraction warnings, speeding, bad driving, mileage, forward collision warnings, offline warnings, pedestrian collision warnings, monthly order volume, number of transport vehicles, and hazardous chemical categories from the dimensions of driving performance, vehicle driving conditions, and hazardous goods to construct a risk assessment indicator set for transportation enterprises.

4. The risk prediction system for transportation enterprises based on hazardous chemical risk profiles as described in claim 2, characterized in that, It also includes a module for quantifying the risks of dangerous goods, which is configured as follows: The sum of the ratios of the actual quantity (online quantity) of various hazardous chemicals within the unit to the critical quantity specified in "Identification of Major Hazard Installations of Hazardous Chemicals" (GB18218), after correction by a correction factor. As a risk coefficient value for hazardous chemicals.

5. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the method described in claim 1.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it performs the steps of the method described in claim 1 above.

Images