Enterprise risk hidden danger management performance evaluation method and system based on Monte Carlo method

Abstract

The invention discloses an enterprise risk hidden danger management performance evaluation method and system based on a Monte Carlo method, and the method comprises the steps: employing a Delphi method to determine and construct an enterprise risk hidden danger management performance evaluation index system, the system comprises four first-level indexes including enterprise risk value analysis, risk assessment completion rate, risk control level and hidden danger management. Collecting, obtaining and calculating each index in the index system; and calculating an enterprise risk hidden danger management performance evaluation result, and performing evaluation grading. According to the method, a more scientific and reasonable index system used for representing the enterprise risk hidden danger management performance is constructed, the triangular fuzzy number coupling Monte Carlo enterprise risk hidden danger management performance evaluation model is established, the enterprise risk hidden danger management performance evaluation value is obtained, meanwhile, the enterprise performance evaluation range value is obtained, and the enterprise performance evaluation efficiency is improved. Quantitative evaluation of enterprise risk hidden danger management performance level is realized, and uncertainty of enterprise risk hidden danger management is reflected to a certain extent.

Description

Technical Field

[0001] The present invention belongs to the technical field of chemical industry safety, and particularly relates to a method and system for evaluating the performance of enterprise risk and hidden danger management based on the Monte Carlo method. Background Art

[0002] The newly revised "Guidelines for Chemical Process Safety Management AQ / T 3034-2022" proposes that enterprises should integrate the risk grading and control and hidden danger investigation and treatment working mechanisms into risk management work. Through hazard identification, risk assessment, risk control and risk monitoring, ensure that risks are under control. Therefore, risk and hidden danger management is also one of the important professional safeties in the process safety management of refining enterprises.

[0003] There are hundreds of refining enterprises in our country. Every year, a large amount of manpower and material resources are invested in inspections and audits to evaluate the risk and hidden danger management level of enterprises and discover their management defects. If the risk and hidden danger management status of enterprises can be judged in real time through a scientific index system, the inspection and audit frequency can be reduced to a certain extent, saving manpower and material costs and improving the economic benefits of enterprises.

[0004] CN112884312A discloses a method for risk assessment of chemical plants based on OWA operator weighting. According to the analysis of chemical risk factors, the first-level risk factor indicators and second-level risk factor indicators for evaluating the risk factors of process plants are obtained, and the decision data sets of risk factor indicators corresponding to each indicator are used to compare and analyze all the risk factors involved, summarize and induce the priority of risk factors affecting existing chemical plants, and conduct qualitative evaluation, and implement it preferentially in actual production. This method mainly analyzes the risk factors of a single chemical plant and relies on expert experience, and does not propose a corresponding index system and evaluation method for the enterprise risk and hidden danger management status.

[0005] CN115713228A discloses a risk analysis method for petrochemical plants. By extracting the safety-related parameters of petrochemical plants and each protection layer associated with the fluctuations of the safety-related parameters, and based on the safety-related parameter operation data in the corresponding protection layer, a risk-related index for characterizing the risk status of the petrochemical plant is calculated. This method conducts risk analysis for specific plants and specific processes, and realizes the prediction of the risk level of petrochemical plants. However, it mainly analyzes the plant process and ignores the safety management level of enterprises.

[0006] CN115511236A discloses a method and device for dynamically evaluating the safety risks in petrochemical processes, which determine the key safety variables in the target petrochemical process and the abnormal events of the key safety variables, analyze the out-of-control consequences based on the abnormal events, and calculate the consequence losses under different accident consequence calculation scenarios; analyze the safety protection layers of the abnormal events of the key safety variables, and calculate the dynamic occurrence probabilities of the abnormal events in different safety protection layers according to the initial failure probabilities and historical data of the different safety protection layers; and calculate the dynamic safety risks of the petrochemical process according to the consequence losses and the dynamic occurrence probabilities of the different safety protection layers. This method mainly evaluates the impact on the occurrence probability and consequence losses by real-time monitoring of the dynamic data of the device, and calculates the process safety risks of the device, but it cannot intuitively show the risk hidden danger management status of the enterprise.

[0007] Most of the existing risk quantification and evaluation methods mentioned above are for the evaluation and calculation methods of single devices, and rely too much on expert experience. There is less research on the performance index system related to the overall risk hidden danger management of enterprises and the complex relationships among the indexes. At the same time, the existing methods can usually accurately calculate the risk evaluation results of the devices, and the calculation results are single real numbers. However, in actual scenarios, the enterprise's risk hidden danger management performance cannot be fully characterized only by this result. There are many evaluation indexes related to the enterprise's risk hidden danger management performance. For different enterprises, their inherent risk levels and historical accident situations are also different. It is difficult to directly and accurately reflect the enterprise's risk hidden danger management performance through a single value.

[0008] Therefore, the present invention proposes a method and system for evaluating the performance of enterprise risk hidden danger management based on the coupling of triangular fuzzy numbers and Monte Carlo simulation. A more scientific and reasonable index system for characterizing the performance of enterprise risk hidden danger management is constructed. The scales of the various indexes are removed, and the calculation standards of different enterprises are unified. At the same time, a data collection module is developed to automatically calculate the index values by collecting fields from the existing system, and determine the index weights and index ratings according to the index values. Using the performance evaluation module, triangular fuzzy numbers are assigned to the indexes according to the enterprise's inherent risks and historical accidents, and the triangular fuzzy numbers are simulated by the Monte Carlo method. An evaluation model for enterprise risk hidden danger management performance based on the coupling of triangular fuzzy numbers and Monte Carlo is established. Finally, the simulation results are sorted in descending order, and the upper and lower limits of the confidence interval and their average value at a confidence level of 95% are taken as the calculation results, so as to obtain both the evaluation value of the enterprise's risk hidden danger management performance and the range value of the enterprise's performance evaluation, which not only realizes the quantitative evaluation of the enterprise's risk hidden danger management performance level, but also reflects the uncertainty of the enterprise's risk hidden danger management to a certain extent. Summary of the Invention

[0009] In view of the problems existing in the above-mentioned prior art, the present invention proposes a performance evaluation method and system for enterprise risk and hidden danger management based on the Monte Carlo method, with reasonable design, which overcomes the deficiencies of the prior art and has good effects.

[0010] In order to achieve the above object 1, the present invention adopts the following technical solutions:

[0011] A performance evaluation method for enterprise risk and hidden danger management based on the Monte Carlo method, comprising the following steps:

[0012] S1. Determine and construct a performance evaluation index system for enterprise risk and hidden danger management by using the Delphi method. This system includes four first-level indicators: enterprise risk value analysis, risk assessment completion rate, risk control level, and hidden danger management.

[0013] S2. Collect, obtain, and calculate each indicator in the index system.

[0014] S3. Calculate the performance evaluation result of enterprise risk and hidden danger management and conduct evaluation grading.

[0015] Further, in S1, the enterprise risk value analysis indicator includes two secondary sub-indicators, namely enterprise characteristic risk value and average device characteristic risk value; the risk assessment completion rate indicator includes three secondary sub-indicators, namely RC-sheet completion rate, HAZOP analysis completion rate, and SIL analysis completion rate; the risk control level indicator includes four secondary sub-indicators, namely the proportion of major risk devices, risk control measure completion rate, risk control measure overdue rate, and risk control measure proposal rate; the hidden danger management indicator includes four secondary sub-indicators, namely hidden danger rectification rate, average total number of hidden dangers in the device, hidden danger rectification overdue rate, and major hidden danger proposal rate.

[0016] Further, in S2, according to the required indicators, obtain the indicator data automatically from the enterprise's dual-prevention management system according to the hierarchy, year, and month.

[0017] Further, S3 includes the following sub-steps:

[0018] S3.1. Calculate the objective weight of each indicator by using the entropy weight method according to the specific values of the indicators.

[0019] S3.2. Determine the triangular fuzzy number and evaluation grade of each indicator.

[0020] S3.3. Calculate the Monte Carlo formula of the triangular fuzzy number.

[0021] S3.4. Conduct Monte Carlo simulation to obtain the result sequence.

[0022] S3.5. Output the result and rating.

[0023] Furthermore, in S3.1, first, standardize the index data; for the extremely large type indices in the index system, the standard transformation formula is:

[0024]

[0025] For the extremely small type indices, the standard transformation formula is:

[0026]

[0027] Among them, by standardizing the secondary indices under each primary index, a standardized matrix can be obtained where n is time and m is the number of primary indices;

[0028] Secondly, calculate the entropy value. The entropy of the j-th secondary index among the m primary indices is:

[0029]

[0030] In the formula,

[0031] Finally, determine the entropy weight of the j-th secondary index as:

[0032]

[0033] In the formula, 0 ≤ w j ≤ 1 and

[0034] Calculate the weights of each secondary index:

[0035] w ij = w i w j

[0036] In the formula, w i is the weight of the i-th primary index.

[0037] Furthermore, in S3.2, according to the specific values of different indices, determine the index evaluation levels and assign corresponding triangular fuzzy numbers; the evaluation levels include excellent, good, medium, and poor; when the evaluation level is excellent, the triangular fuzzy number is (0, 0.15, 0.3); when the evaluation level is good, the triangular fuzzy number is (0.3, 0.45, 0.6); when the evaluation level is medium, the triangular fuzzy number is (0.6, 0.75, 0.9); when the evaluation level is poor, the triangular fuzzy number is (0.9, 0.95, 1).

[0038] Furthermore, in S3.3, first, construct a triangular fuzzy variable with three real numbers a, b, and c Its membership function It is expressed by the following formula:

[0039]

[0040] Convert the triangular membership function into a triangular fuzzy number The probability density distribution function of the possibility variable is as follows:

[0041]

[0042] After integrating the above probability density function into a probability distribution function, the random simulation formula for the possible value x of the secondary index is obtained by the inverse transformation method:

[0043]

[0044] Among them, u is a random number uniformly distributed on the interval [0, 1].

[0045] Furthermore, in the above S3.4, through Monte Carlo simulation, a series of uniformly distributed random numbers u1, u2,..., u n on the interval [0, 1] are generated and substituted into the random simulation formula to obtain a series of random simulation values x1, x2,..., x n of the single index. And after enough calculation test times, the possible value sequences of each index are obtained; the enterprise risk and hidden danger management performance evaluation result sequence is determined; the single simulation evaluation result formula is:

[0046] S l = w ij x ij

[0047] Among them, w ij is the weight of each index, x ij is the Monte Carlo simulation value of each index, and S l is the l-th simulation value of the enterprise risk and hidden danger management performance evaluation result, l ∈ (1, N), and N is the number of Monte Carlo simulations.

[0048] Furthermore, in the above S3.5, the simulated series result sequence {S l |l = 1 ~ N} is arranged from large to small. According to the empirical cumulative frequency formula:

[0049] P l = l / (N + 1), l = 1 ~ N

[0050] Among them, P l is the empirical cumulative frequency of S l corresponding to the serial number l after descending order;

[0051] Construct the confidence interval of the enterprise risk and hidden danger management performance level at the confidence level ɑ as:

[0052] [S INT((1-0.5(1-α))(N+1)) ,S INT(0.5(1-α)(N+1))

[0053] For a single indicator, usually when the number of simulation times N = 10,000, the result converges. When the confidence level reaches 95%, the final calculation result is determined, and the interval distribution of the enterprise risk and hidden danger management performance level is as follows:

[0054] [S 9751 ,S 250

[0055] At the same time, the final result is determined as:

[0056]

[0057] The enterprise risk and hidden danger management performance level is classified. When S ∈ (0, 0.3), the evaluation level is excellent; when S ∈ (0.3, 0.6), the evaluation level is good; when S ∈ (0.6, 0.9), the evaluation level is medium; when S ∈ (0.9, 1), the evaluation level is poor.

[0058] To achieve the above-mentioned purpose 2, the present invention adopts the following technical solutions:

[0059] An enterprise risk and hidden danger management performance evaluation system based on the Monte Carlo method, including a risk and hidden danger management index library module, an index automatic collection module, and a risk and hidden danger management performance evaluation module;

[0060] The risk and hidden danger management index library module is used to determine and construct an enterprise risk and hidden danger management performance evaluation index system based on the connotation and goal of enterprise risk and hidden danger evaluation, based on the protection layer theory, combined with the enterprise risk and hidden danger management scenario and the actual management situation, by using the Delphi method;

[0061] The index automatic collection module is used to realize the automatic collection, acquisition, and calculation of each index in the index system. According to the indexes required by the performance evaluation system, it automatically obtains data from the enterprise's dual-prevention management system according to the hierarchy, year, and month, stores this data in the form of a file, and after verifying the validity of the data, performs an operation of storing the valid data in the database;

[0062] The risk and hidden danger management performance evaluation module is used to calculate the enterprise risk and hidden danger management performance evaluation result and perform evaluation classification.

[0063] The beneficial technical effects brought by the present invention:

[0064] ​​1. Currently, there is no relevant index system and evaluation criteria for the risk and potential hazard management performance evaluation of enterprises. Based on the enterprise business scenario and the protection layer theory, the present invention establishes an index system related to risk and potential hazard management, clarifies relevant first-level and second-level indicators, and selects de-scaling indicators, thus establishing evaluation criteria for the risk and potential hazard management performance of different enterprises.

[0065] 2. The existing inventions mainly focus on the risk assessment of a single device, and the calculation results are usually a certain value. Although the problem of risk level classification is solved, there are many uncertain factors in the risk and potential hazard management performance evaluation of enterprises. Usually, a single value cannot represent its management complexity. The present invention proposes a calculation method based on the coupling of triangular fuzzy numbers and Monte Carlo simulation, which uses probability statistics methods to characterize the uncertainty of enterprise risk and potential hazard management. It can not only quantitatively calculate the evaluation value of enterprise risk and potential hazard management performance, but also obtain the interval value of enterprise risk and potential hazard management performance evaluation. Enterprise managers can appropriately strengthen the risk and potential hazard management of the enterprise according to the upper and lower limits of the result interval, which helps the enterprise to further improve the level of risk and potential hazard management.

[0066] 3. Currently, there is no clear qualitative and quantitative calculation method for the overall risk and potential hazard management status of enterprises. The present invention provides a method for objectively calculating the index weights, and combines triangular fuzzy distribution and Monte Carlo simulation to realize the performance evaluation grading of each index and enterprise risk and potential hazard management. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 It is the flow chart of the enterprise risk and potential hazard management performance evaluation method in the present invention;

[0068] Figure 2 It is the schematic diagram of the enterprise risk and potential hazard management performance evaluation index system in the present invention;

[0069] Figure 3 It is the membership function diagram of the possible value variable of the triangular fuzzy number in the present invention;

[0070] Figure 4 It is the probability density distribution function diagram of the triangular fuzzy number in the present invention;

[0071] Figure 5 It is the schematic diagram of the enterprise risk and potential hazard management performance evaluation system in the present invention; DETAILED DESCRIPTION OF THE INVENTION

[0072] The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. On the contrary, they are merely examples of methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0073] The terms used in this disclosure are for the purpose of describing specific embodiments only and are not intended to limit the disclosure. The singular forms "a", "the", and "said" used in this disclosure and the appended claims are also intended to include the plural forms unless the context clearly dictates otherwise. It should also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0074] The following further illustrates the specific implementation manners of the present invention in conjunction with specific embodiments:

[0075] An enterprise risk and potential hazard management performance evaluation method based on the Monte Carlo method, as Figure 1 shown, includes the following steps:

[0076] S1. Use the Delphi method to determine and construct an enterprise risk and potential hazard management performance evaluation index system, as Figure 2 shown. This system includes a total of four first-level indicators: enterprise risk value analysis, risk assessment completion rate, risk control level, and potential hazard management.

[0077] The enterprise risk value analysis indicator includes two second-level sub-indicators, namely enterprise characteristic risk value and average device characteristic risk value; the risk assessment completion rate indicator includes three second-level sub-indicators, namely RC-sheet completion rate, HAZOP analysis completion rate, and SIL analysis completion rate; the risk control level indicator includes four second-level sub-indicators, namely the proportion of major risk devices, risk control measure completion rate, risk control measure overdue rate, and risk control measure proposal rate; the potential hazard management indicator includes four second-level sub-indicators, namely potential hazard rectification rate, average total number of potential hazards in the device, potential hazard rectification overdue rate, and major potential hazard proposal rate. The characteristics of the above indicators are all easy to quantify and de-scale, facilitating subsequent calculation and comparison of the risk and potential hazard management performance of different enterprises.

[0078] S2. Collect, obtain, and calculate each indicator in the indicator system;

[0079] According to the required indicators, automatically obtain the indicator data from the enterprise's dual-prevention management system based on the hierarchy, year, and month.

[0080] S3. Calculate the evaluation results of enterprise risk and potential hazard management performance and conduct evaluation grading. First, according to the specific values of the indicators, use the entropy weight method to calculate the objective weights of each indicator. Then, based on the triangular fuzzy numbers assigned to the indicator ratings, use the Monte Carlo method to simulate the possible value variables of each indicator N times, and perform weighted summation on the results of each simulation to obtain N simulated evaluation values. Sort the final results in descending order, and take the average value of the confidence interval with a confidence level of 95%. This average value can be used as the evaluation result of the enterprise's risk and potential hazard management performance. At the same time, since the evaluation results show an interval distribution, the upper and lower limits also reflect the uncertainty of the enterprise's risk and potential hazard management. For enterprises with lower lower limits, risk and potential hazard management should be strengthened. Specifically, it includes the following sub-steps:

[0081] S3.1. According to the specific values of the indicators, use the entropy weight method to calculate the objective weights of each indicator;

[0082] In the above-mentioned S3.1, first, standardize the indicator data; for the extremely large-type indicators in the indicator system, the standard transformation formula is:

[0083]

[0084] For the extremely small-type indicators, the standard transformation formula is:

[0085]

[0086] In this indicator system, the enterprise characteristic risk value, average device characteristic risk value, proportion of major risk devices, overdue rate of risk control measures, average total number of potential hazards in the device, and overdue rate of potential hazard rectification are extremely small-type indicators, and the rest of the indicators are extremely large-type indicators. Standardize the secondary indicators under each first-level indicator to obtain the standardized matrix where n is the time (month / year) and m is the number of first-level indicators;

[0087] Secondly, calculate the entropy value. The entropy of the j-th secondary indicator among the m first-level indicators is:

[0088]

[0089] In the formula,

[0090] Finally, determine the entropy weight of the j-th secondary indicator as:

[0091]

[0092] In the formula, 0 ≤ w j ≤ 1 and

[0093] Calculate the weights of each secondary indicator:

[0094] w ij = w i w j

[0095] where w i is the weight of the i-th first-level index.

[0096] S3.2. Determine the triangular fuzzy numbers and evaluation grades of each index;

[0097] According to the specific values of different indexes, determine the index evaluation grades and assign corresponding triangular fuzzy numbers; here, the triangular fuzzy numbers can be adjusted according to the inherent risk level of the enterprise, the historical risk hidden danger management situation, and historical accidents. The evaluation grades include excellent, good, medium, and poor; when the evaluation grade is excellent, the triangular fuzzy number is (0, 0.15, 0.3); when the evaluation grade is good, the triangular fuzzy number is (0.3, 0.45, 0.6); when the evaluation grade is medium, the triangular fuzzy number is (0.6, 0.75, 0.9); when the evaluation grade is poor, the triangular fuzzy number is (0.9, 0.95, 1). As shown in the following table:

[0098] Evaluation level Triangular fuzzy number Excellent (0，0.15，0.3) Good (0.3，0.45，0.6) Medium (0.6，0.75，0.9) Poor (0.9，0.95，1)

[0099] S3.3. Calculate the stochastic simulation formula of the triangular fuzzy number;

[0100] First, construct a triangular fuzzy variable with three real numbers a, b, and c Its membership function is expressed by the following formula:

[0101]

[0102] Convert the triangular membership function into the probability density distribution function of the possibility variable of the triangular fuzzy number The formula is as follows:

[0103]

[0104] After integrating the above probability density function into the probability distribution function, use the inverse transformation method to obtain the stochastic simulation formula of the possible value x of the secondary index:

[0105]

[0106] where u is a random number uniformly distributed on the interval [0, 1].

[0107] S3.4. Conduct Monte Carlo simulation to obtain the result sequence;

[0108] Through Monte Carlo simulation, generate a series of uniformly distributed random numbers u1, u2,..., u on the interval [0, 1] n, substitute it into the random simulation formula to obtain a series of random simulation values \(x_1, x_2, \cdots, x\) of a single index. n , use the operations between real numbers to replace the operations between triangular fuzzy numbers, and after enough calculation test times, obtain the possible value sequences of each index; determine the sequence of enterprise risk and hidden danger management performance evaluation results; the formula for the single simulation evaluation result is:

[0109] S l = w ij x ij

[0110] where \(w\) ij is the weight of each index, \(x\) ij is the Monte Carlo simulation value of each index, \(S\) l is the \(l\)-th simulation value of the enterprise risk and hidden danger management performance evaluation result, \(l\in(1, N)\), \(N\) is the number of Monte Carlo simulations. The more the number of simulations \(N\), the closer the distribution of the \(S\) sequence is to the true distribution. In fact, generally take the number of simulations when the test results converge.

[0111] S3.5. Output results and ratings;

[0112] Arrange the simulated series result sequence \(\{S\) l |l = 1\sim N\}\) from large to small. According to the empirical cumulative frequency formula:

[0113] P l = l / (N + 1), l = 1\sim N

[0114] where \(P\) l is the empirical cumulative frequency of \(S\) l corresponding to the serial number \(l\) after descending order;

[0115] Construct the confidence interval of the enterprise risk and hidden danger management performance level at the confidence level \(\alpha\) as:

[0116] [S INT((1-0.5(1-α))(N+1)) , S INT(0.5(1-α)(N+1))

[0117] For a single index, usually when the number of simulations \(N = 10000\) times, the results converge. Determine the final calculation results with a confidence level of 95%. Then the interval distribution of the enterprise risk and hidden danger management performance level is:

[0118] [S 9751 , S 250

[0119] At the same time, determine the final result as:

[0120]

[0121] ​​The performance level of enterprise risk and potential hazard management is graded. When S ∈ (0, 0.3), the evaluation level is excellent. When S ∈

[0122] (0.3, 0.6), the evaluation level is good. When S ∈ (0.6, 0.9), the evaluation level is medium. When S ∈ (0.9, 1), the evaluation level is poor, as shown in the following table:

[0123] S Evaluation level (0，0.3) Excellent (0.3，0.6) Good (0.6，0.9) Medium (0.9，1) Poor

[0124] Through this method, the enterprise risk and potential hazard management level can be obtained as S, and its possible distribution interval is [S 9751 , S 250 . If the upper limit S of the enterprise interval 250 is relatively high, it indicates that its inherent risk is relatively high or there are many historical accidents. Even if the grade where the S value is located is excellent or good, the risk management level should be improved.

[0125] An enterprise risk and potential hazard management performance evaluation system based on the Monte Carlo method, as Figure 5 shown, includes a risk and potential hazard management index library module, an index automatic collection module, and a risk and potential hazard management performance evaluation module;

[0126] The risk and potential hazard management index library module is used to determine and construct an enterprise risk and potential hazard management performance evaluation index system based on the connotation and objectives of enterprise risk and potential hazard evaluation, based on the protection layer theory, combined with the enterprise risk and potential hazard management scenario and actual management situation, using the Delphi method;

[0127] The index automatic collection module is used to automatically collect, obtain, and calculate each index in the index system. According to the indexes required by the performance evaluation system, it automatically obtains data from the enterprise's dual prevention management system according to the affiliated level, year, and month, stores this data in the form of a file, and after verifying the validity of the data, performs an operation to store the valid data in the database;

[0128] The risk and potential hazard management performance evaluation module is used to calculate the enterprise risk and potential hazard management performance evaluation result and conduct evaluation grading.

[0129] Example 1

[0130] An enterprise risk and potential hazard management performance evaluation method based on the Monte Carlo method includes the following steps:

[0131] S1. Use the Delphi method to determine and construct an enterprise risk and potential hazard management performance evaluation index system, which includes four first-level indexes: enterprise risk value analysis, risk assessment completion rate, risk control level, and potential hazard management;

[0132] The enterprise risk value analysis indicators include two secondary sub - indicators, namely the enterprise characteristic risk value and the average device characteristic risk value; the risk assessment completion rate indicator includes three secondary sub - indicators, namely the RC - sheet completion rate, the HAZOP analysis completion rate, and the SIL analysis completion rate; the risk control level indicator includes four secondary sub - indicators, namely the proportion of major risk devices, the risk control measure completion rate, the overdue rate of risk control measures, and the proposed rate of risk control measures; the hidden danger management indicator includes four secondary sub - indicators, namely the hidden danger rectification rate, the average total number of hidden dangers in the device, the overdue rate of hidden danger rectification, and the proposed rate of major hidden dangers. The characteristics of the above indicators are all easy to quantify and de - scale, which is convenient for subsequent calculation and comparison of the risk and hidden danger management performance of different enterprises.

[0133] S2. Collect, obtain, and calculate each indicator within the indicator system;

[0134] According to the required indicators, automatically obtain the indicator data from the enterprise's dual - prevention management system according to the affiliated level, year, and month.

[0135] S3. Calculate the evaluation result of the enterprise risk and hidden danger management performance and conduct evaluation grading. First, according to the specific values of the indicators, use the entropy weight method to calculate the objective weights of each indicator. Then, according to the triangular fuzzy numbers assigned to the indicator ratings, use the Monte Carlo method to simulate the possible value variables of each indicator N times, and perform weighted summation on each simulation result to obtain N simulation evaluation values. Arrange the final results in descending order, and take the average value of the confidence interval with a confidence level of 95%. This average value can be used as the evaluation result of the enterprise's risk and hidden danger management performance. At the same time, since the evaluation results show an interval distribution, the upper and lower limits also reflect the uncertainty of the enterprise's risk and hidden danger management. For enterprises with a lower lower limit, risk and hidden danger management should be strengthened. It specifically includes the following sub - steps:

[0136] S3.1. To eliminate the influence of human factors, according to the specific values of the indicators, use the entropy weight method to calculate the objective weights of each indicator;

[0137] In the above S3.1, first, standardize the indicator data; for the extremely large - type indicators in the indicator system, the standard transformation formula is:

[0138]

[0139] For the extremely small - type indicators, the standard transformation formula is:

[0140]

[0141] Among them, after standardizing each secondary indicator under each first - level indicator, a standardized matrix can be obtained Among them, n is the time, and m is the number of first - level indicators;

[0142] Secondly, calculate the entropy value. The entropy of the j-th secondary index among the m primary indices is as follows:

[0143]

[0144] In the formula,

[0145] Finally, determine the entropy weight of the j-th secondary index as:

[0146]

[0147] In the formula, 0 ≤ w k ≤ 1 and

[0148] Calculate the weights of each secondary index for the target layer:

[0149] w ij = w i w j

[0150] In the formula, w i is the weight of each primary index. There are four primary indices in total, and each is weighted with 0.25 to calculate the weights of each secondary index.

[0151] S3.2. Determine the triangular fuzzy numbers and evaluation grades of each index;

[0152] According to the specific values of different indices, determine the index evaluation grades and assign corresponding triangular fuzzy numbers; here, the triangular fuzzy numbers can be adjusted according to the inherent risk level of the enterprise, the historical risk hazard management situation, and historical accidents. The evaluation grades include excellent, good, medium, and poor; when the evaluation grade is excellent, the triangular fuzzy number is (0, 0.15, 0.3); when the evaluation grade is good, the triangular fuzzy number is (0.3, 0.45, 0.6); when the evaluation grade is medium, the triangular fuzzy number is (0.6, 0.75, 0.9); when the evaluation grade is poor, the triangular fuzzy number is (0.9, 0.95, 1).

[0153] Combined with the risk matrix theory, the enterprise characteristic risk value and the average characteristic risk value of the device, the index values and their corresponding evaluation grades and triangular fuzzy numbers are shown in the table:

[0154] Index value Evaluation level Triangular fuzzy number [1,10) Excellent (0，0.15，0.3) [10,20) Good (0.3，0.45，0.6) [20,40) Medium (0.6，0.75，0.9) [40,200] Poor (0.9，0.95，1)

[0155] In addition, after the regularization in step 3.1, the ranges of the values of the remaining secondary indices are all 0 - 100%. Combined with the existing data and the actual situation of the enterprise, the index values and their corresponding evaluation grades and triangular fuzzy numbers are shown in the table:

[0156] Index value (%) Evaluation level Triangular fuzzy number (80,100] Excellent (0，0.15，0.3) (60,80] Good (0.3，0.45，0.6) (40,60] Medium (0.6，0.75，0.9) [0,40] Poor (0.9，0.95，1)

[0157] S3.3. Calculate the stochastic simulation formula of triangular fuzzy numbers;

[0158] First, construct a triangular fuzzy variable with three real numbers a, b, and c Its membership function Is expressed by the following formula:

[0159]

[0160] Obviously, x is the possible value variable in the triangular fuzzy number;

[0161] The membership function is as Figure 3 Shown, The change interval of is [0, 1];

[0162] Dividing the membership function by the area enclosed by the membership function and the x-axis can obtain the value of the membership function per unit area, which can also be used as the probability density distribution function of the possibility variable of the triangular fuzzy number The formula is as follows:

[0163]

[0164] The triangular fuzzy number The image of the probability density distribution function is as Figure 4 Shown;

[0165] For a single-dimensional random variable x, if its probability distribution function is F(x) and its probability density function is f(x), and f(x) is continuous in x, then the relationship between the two is as follows:

[0166] f(x) = dF(x) / dx

[0167] Therefore, after converting the above probability density function integral into a probability distribution function and using the inverse transformation method, the stochastic simulation formula for the possible value x of the secondary index can be obtained:

[0168]

[0169] Among them, u is a random number uniformly distributed on the interval [0, 1].

[0170] Convert the triangular membership function into the probability density distribution function of the possibility variable of the triangular fuzzy number The formula is as follows:

[0171]

[0172] After converting the above probability density function integral into a probability distribution function and using the inverse transformation method, the stochastic simulation formula for the possible value x of the secondary index can be obtained:

[0173]

[0174] Among them, u is a random number uniformly distributed on the interval [0, 1].

[0175] S3.4. Conduct Monte Carlo simulation to obtain the result sequence;

[0176] Through Monte Carlo simulation, a series of uniformly distributed random numbers u1, u2,..., u on the interval [0, 1] are generated n , and substituting them into the formula, a series of random simulation values x1, x2,..., x of a single index are obtained n , and after a sufficient number of calculation test times, the possible value sequences of each index are obtained; determine the result sequence of enterprise risk and hidden danger management performance evaluation; the formula for the single simulation evaluation result is:

[0177] S l = w ij x ij

[0178] Among them, w ij is the weight of each index, x ij is the Monte Carlo simulation value of each index, and S l is the l-th simulation value of the enterprise risk and hidden danger management performance evaluation result, l ∈ (1, N), N is the number of Monte Carlo simulations. The more the number of simulations N, the closer the distribution of the S sequence is to the true distribution. In fact, generally, the number of simulations when the test results converge is taken.

[0179] S3.5. Output the results and ratings.

[0180] Arrange the simulated series result sequence {S l |l = 1 ~ N} from largest to smallest. According to the empirical cumulative frequency formula:

[0181] P l = l / (N + 1), l = 1 ~ N

[0182] Among them, P l is the empirical cumulative frequency of S l corresponding to the serial number l after descending order;

[0183] Construct the confidence interval of the enterprise risk and hidden danger management performance level at the confidence level ɑ as:

[0184] [S INT((1-0.5(1-α))(N+1)) , S INT(0.5(1-α)(N+1))

[0185] For a single index, usually when the number of simulations N = 10000 times, the results converge. With the confidence level reaching 95%, the final calculation results are determined. Then the interval distribution of the enterprise risk and hidden danger management performance level is:

[0186] [S​9751 , S 250

[0187] Meanwhile, it is determined that the final result is:

[0188]

[0189] The performance level of enterprise risk and potential hazard management is graded. When S ∈ (0, 0.3), the evaluation level is excellent; when S ∈ (0.3, 0.6), the evaluation level is good; when S ∈ (0.6, 0.9), the evaluation level is medium; when S ∈ (0.9, 1), the evaluation level is poor, as shown in the following table:

[0190] S Evaluation level (0，0.3) Excellent (0.3，0.6) Good (0.6，0.9) Medium (0.9，1) Poor

[0191] Through this method, the enterprise risk and potential hazard management level S can be obtained, and its possible distribution interval is [S 9751 , S 250 . If the upper limit S of the enterprise interval 250 is relatively high, it indicates that its inherent risk is relatively high or there are many historical accidents. Even if the level where the S value is located is excellent or good, the risk management level should still be improved.

[0192] The above describes the preferred embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and the equipment and structures not described in detail should be understood to be implemented in a common manner in the art; any person skilled in the art can, without departing from the scope of the technical solution of the present invention, make many possible changes and modifications to the technical solution of the present invention by using the methods and technical contents disclosed above, or modify it into equivalent embodiments with equivalent changes, which does not affect the essence of the present invention. Therefore, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of protection of the technical solution of the present invention.​

Claims

1. An enterprise risk and potential hazard management performance evaluation method based on the Monte Carlo method, characterized in that It includes the following steps: S1. Determine and construct an enterprise risk and potential hazard management performance evaluation index system by using the Delphi method. This system includes four first-level indicators: enterprise risk value analysis, risk assessment completion rate, risk control level, and potential hazard management. S2. Collect, obtain, and calculate each indicator within the index system. S3. Calculate the enterprise risk and potential hazard management performance evaluation results and conduct evaluation grading.

2. The enterprise risk hidden danger management performance evaluation method based on the Monte Carlo method according to claim 1, characterized in that In S1, the enterprise risk value analysis indicator includes two second-level sub-indicators, namely enterprise characteristic risk value and average device characteristic risk value. The risk assessment completion rate indicator includes three second-level sub-indicators, namely RC-sheet completion rate, HAZOP analysis completion rate, and SIL analysis completion rate; the risk control level indicator includes four second-level sub-indicators, namely the proportion of major risk devices, risk control measure completion rate, risk control measure overdue rate, and risk control measure proposal rate; the potential hazard management indicator includes four second-level sub-indicators, namely potential hazard rectification rate, average total number of potential hazards in the device, potential hazard rectification overdue rate, and major potential hazard proposal rate.

3. The performance evaluation method for enterprise risk and potential hazard management based on the Monte Carlo method according to claim 1, wherein In S2, according to the required indicators, obtain the indicator data automatically from the enterprise's dual prevention management system based on the affiliated level, year, and month.

4. A performance evaluation method for enterprise risk and potential hazard management based on the Monte Carlo method according to claim 1, characterized in that S3 includes the following sub-steps: S3.

1. Calculate the objective weights of each indicator by using the entropy weight method according to the specific values of the indicators. S3.

2. Determine the triangular fuzzy numbers and evaluation grades of each indicator. S3.

3. Calculate the stochastic simulation formula of the triangular fuzzy numbers. S3.

4. Conduct Monte Carlo simulation to obtain the result sequence. S3.

5. Output the results and ratings.

5. The performance evaluation method for enterprise risk and potential hazard management based on the Monte Carlo method according to claim 4, characterized in that, In S3.1, first, standardize the indicator data; for the extremely large-type indicators in the indicator system, its standard transformation formula is: For the extremely small-type indicators, its standard transformation formula is: Among them, by standardizing the secondary indicators under each primary indicator, a standardized matrix can be obtained where n is the time and m is the number of primary indicators; Secondly, calculate the entropy value. The entropy of the j-th second-level indicator among m first-level indicators is: In the formula, Finally, determine the entropy weight of the j-th second-level indicator as: where 0 ≤ w j ≤ 1 and Calculate the weights of each second-level indicator: w ij = w i w j where w i is the weight of the i-th first-level index.

6. The performance evaluation method for enterprise risk and potential hazard management based on the Monte Carlo method according to claim 4, wherein In S3.2, according to the specific values of different indicators, determine the indicator evaluation grades and assign corresponding triangular fuzzy numbers; the evaluation grades include excellent, good, medium, and poor; when the evaluation grade is excellent, the triangular fuzzy number is (0, 0.15, 0.3); when the evaluation grade is good, the triangular fuzzy number is (0.3, 0.45, 0.6); when the evaluation grade is medium, the triangular fuzzy number is (0.6, 0.75, 0.9); when the evaluation grade is poor, the triangular fuzzy number is (0.9, 0.95, 1).

7. A performance evaluation method for enterprise risk and potential hazard management based on the Monte Carlo method according to claim 4, characterized in that In the step S3.3, first, a triangular fuzzy variable is constructed with three real numbers a, b, and c. Its membership function is expressed by the following formula: Convert the triangular membership function into a triangular fuzzy number The probability density distribution function of the possibility variable is as follows: After converting the above probability density function integral into a probability distribution function, use the inverse transformation method to obtain the stochastic simulation formula of the possible value x of the second-level indicator: where u is a random number uniformly distributed on the interval [0, 1].

8. A performance evaluation method for enterprise risk and potential hazard management based on the Monte Carlo method according to claim 4, characterized in that, In S3.4, a series of uniformly distributed random numbers u1, u2,..., u in the interval [0, 1] are generated through Monte Carlo simulation n , which are substituted into the random simulation formula to obtain a series of random simulation values x1, x2,..., x of a single index n , and after a sufficient number of calculation test times, sequences of possible values of each index are obtained; a sequence of evaluation results for enterprise risk and hidden danger management performance is determined; the formula for a single simulation evaluation result is: S l = w ij x ij Among them, w ij is the weight of each index, x ij is the Monte Carlo simulation value of each index, and S l is the l-th simulation value of the enterprise risk and hidden danger management performance evaluation result, where l ∈ (1, N) and N is the number of Monte Carlo simulations.

9. The performance evaluation method for enterprise risk and potential hazard management based on the Monte Carlo method according to claim 4, wherein In S3.5, the simulated series result sequence {S l |l = 1 to N} is sorted from largest to smallest. According to the empirical cumulative frequency formula: P l = l / (N + 1), l = 1 to N Among them, P l is the experience accumulation frequency of S corresponding to the serial number l after descending order arrangement l ; Construct the confidence interval of the enterprise risk and potential hazard management performance level at the confidence level ɑ as: [S INT((1-0.5(1-α))(N+1)) ,S INT(0.5(1-α)(N+1)) ​ For a single indicator, usually when the number of simulation times N = 10000 times, the results converge. Determine the final calculation results with a confidence level of 95%. Then the interval distribution of the enterprise risk and potential hazard management performance level is: [S 9751 ,S 250 ​ At the same time, determine the final result as: The performance level of enterprise risk and potential hazard management is graded. When S ∈ (0, 0.3), the evaluation level is excellent. When S ∈ (0.3, 0.6), the evaluation level is good. When S ∈ (0.6, 0.9), the evaluation level is medium. When S ∈ (0.9, 1), the evaluation level is poor.

10. An enterprise risk and potential hazard management performance evaluation system based on the Monte Carlo method, characterized in that, It includes a risk and potential hazard management index library module, an index automatic collection module, and a risk and potential hazard management performance evaluation module; The risk and potential hazard management index library module is used to determine and construct an enterprise risk and potential hazard management performance evaluation index system by using the Delphi method based on the connotation and objectives of enterprise risk and potential hazard evaluation, the protection layer theory, and combining the enterprise risk and potential hazard management scenario and the actual management situation; The index automatic collection module is used to automatically collect, obtain, and calculate each index in the index system. According to the indexes required by the performance evaluation system, it automatically obtains data from the enterprise's dual prevention management system according to the hierarchy, year, and month, stores this data in the form of a file, and after verifying the validity of the data, performs an operation to store the valid data in the database; The risk and potential hazard management performance evaluation module is used to calculate the enterprise risk and potential hazard management performance evaluation result and conduct evaluation grading.

Images