CN113065771B - Chemical enterprise accident risk assessment method and system based on index weight optimization - Google Patents
Chemical enterprise accident risk assessment method and system based on index weight optimization Download PDFInfo
- Publication number
- CN113065771B CN113065771B CN202110366696.XA CN202110366696A CN113065771B CN 113065771 B CN113065771 B CN 113065771B CN 202110366696 A CN202110366696 A CN 202110366696A CN 113065771 B CN113065771 B CN 113065771B
- Authority
- CN
- China
- Prior art keywords
- accident
- index
- evaluation
- case
- accident case
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a chemical enterprise accident risk assessment method and system based on index weight optimization, wherein a bipartite graph of accident cases and evaluation indexes is constructed based on accident case data and evaluation indexes of chemical enterprises; calculating the correlation strength of the accident node and the evaluation index by using a recommendation algorithm based on matrix decomposition based on a bipartite graph of the accident case and the evaluation index; calculating the influence of the accident node; based on the correlation strength between the accident node and the historical evaluation indexes and the influence of the accident node, processing the bipartite graph by adopting a random walk algorithm to obtain an optimized weight of each evaluation index; quantitatively evaluating the chemical enterprises by adopting the optimized evaluation indexes, and calculating the accident risk level of the chemical enterprises to be evaluated according to the evaluation score of each index to obtain the potential safety hazard in the safety production of the chemical enterprises; and sending the accident risk level and the potential safety hazard of the chemical enterprise to be evaluated to the mobile terminal of the corresponding manager. The quality and efficiency of safety management work are improved.
Description
Technical Field
The invention relates to the technical field of safety assessment of chemical enterprises, in particular to a chemical enterprise accident risk assessment method and system based on index weight optimization.
Background
The statements in this section merely provide background information related to the present disclosure and may not constitute prior art.
In the face of frequent chemical industry serious accidents, establishing a scientific and effective chemical industry enterprise safety management index system is an important guarantee for ensuring the safety production of chemical industry enterprises and reducing the accident occurrence probability. The safety quantitative evaluation is an important component for evaluating a safety management system and is also an important means for evaluating the safety production management level of enterprises in the industry. The reasonable safety evaluation system not only can accurately evaluate the safety production management level of the chemical enterprises and discover potential safety hazards of enterprise production, but also can provide improvement measures according to the current situation of enterprise production management and reduce potential expectations of accidents.
Whether a security assessment system is reasonable depends mainly on two factors: evaluation index and index weight.
The evaluation indexes are scoring elements in an evaluation system, are potential safety management hazards possibly existing in the evaluation production process, and are incentive sets for generating accident risks.
The index weight reflects the importance of the evaluation index in the occurrence of the accident risk. The greater the weight of the index, the greater the likelihood that an accident will be caused by the index. Whether the index weight is reasonable or not has a great influence on the correctness of the safety assessment. Some evaluation indexes with unreasonable weights can enable enterprises to neglect some potential risk factors in the production process during evaluation, and potential safety hazards are caused.
In addition, after the evaluation index weight is determined, along with the use of an evaluation system, the evaluation index weight is guided by an industry management level and a regional policy to be changed, and the evaluation index weight needs to be dynamically adjusted and optimized, so that the evaluation system can objectively and truly evaluate the production management and risk prevention level.
At present, the determination of index item weight of a safety management index system of a chemical enterprise exposes a plurality of defects in the practical practice and application process, and has the problems of strong subjectivity, difficult determination of weight, difficult convincing, incapability of well guiding the practical production practice and the like.
The risk grade assessment index system research of the dangerous chemical process of petrochemical enterprises [ J ] the science and technology of safety production in China, 2011(10) 91-94. the risk grade assessment of the dangerous chemical process is carried out on the petrochemical enterprises, and the importance assessment index with typical representative significance is selected as a research object according to influence factors related to the characterization of the dangerous chemical process and the actual situation of the chemical process. The indexes are screened by adopting a theoretical analysis method and an expert consultation method, and the weight of the indexes is calculated by adopting an Analytic Hierarchy Process (AHP) combining qualitative analysis and quantitative analysis. However, when the indexes are excessive and the data volume is large, the method cannot well determine the weight, has less quantitative data and more qualitative components, and cannot comprehensively and accurately evaluate the dangerous chemical process risk level of the petrochemical enterprise.
The method comprises the steps of Chenchao, Wangyong, Lingyuan, correction of the weight of a safety management evaluation index of a petrochemical enterprise based on an accident sample entropy weight method [ J ] Shandong chemical industry, 2019,048(004), 202 plus 205. on the basis of safety management evaluation of the petrochemical enterprise, as the traditional weight determining methods such as expert scoring and the like have strong subjectivity, in order to reduce the influence on the quantitative evaluation effect of the safety management level, the weight of the petrochemical enterprise is corrected through the accident sample entropy weight method, and suggestions are provided for the weight of each element of a safety management evaluation system through correlation analysis between an accident sample and safety management elements and quantitative calculation of the entropy weight method. The entropy weight method establishes a corresponding security management evaluation system model after weight modification. However, the subjective intention of decision makers is often neglected easily by the evaluation method, the transverse comparison among all indexes is lacked, the weight of the indexes is limited by the number of samples, and the accuracy of the accident risk evaluation of chemical enterprises is relatively low.
The inventor finds that the prior art has the following technical problems:
the existing index weight quantification method for the accident risk safety assessment of chemical enterprises is mainly a grading optimization method based on expert subjective experience grading, such as an analytic hierarchy process, an entropy weight method and the like. Due to the lack of scoring basis and the difference of knowledge and experience of experts, the scoring optimization method based on the subjective experience of experts is difficult to form accurate evaluation index weight. The existing accident risk safety assessment index weight of the chemical enterprise is not closely related to the current safety production management level, the process level and the technical level of the enterprise, so that the hidden danger of the current safety production system of the chemical enterprise can not be reflected, and the index weight can not form corresponding change along with the improvement of the production level of the chemical enterprise and the updating of the technology.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a chemical enterprise accident risk assessment method and system based on index weight optimization; firstly, a bipartite graph model between an accident case and an evaluation index is constructed according to the incidence relation between the accident root cause and the evaluation index item of a safety evaluation system determined by accident analysts. And then, setting and optimizing the weight of the two types of nodes in the bipartite graph model, and taking the problems of individual difference, incomplete analysis and the like of analysts into consideration, performing supplementary evaluation on the associated importance of the accident case and the evaluation index bipartite graph by adopting a recommendation algorithm based on matrix decomposition to calculate the association strength between the evaluation index node and the accident node, and calculating the influence of the accident node by synthesizing the occurrence time of the accident case sample and the hazard degree of the accident. Then, an index weight increment optimization algorithm based on the Personalrank random walk is designed, and the weight of the index item is adjusted under the weight constraint of an index system. Therefore, a weight result which is relatively objective for the index system of the safety quantitative evaluation of the chemical enterprises is obtained. Finally, an evaluation index weight increment optimization system is designed, the modeling method is integrated into the system, the increment optimization work of the evaluation index system weight can be completed more visually and conveniently, and the rapidness and the traceability of the evaluation index system weight optimization work are effectively improved.
In a first aspect, the invention provides a chemical enterprise accident risk assessment method based on index weight optimization;
the chemical enterprise accident risk assessment method based on index weight optimization comprises the following steps:
constructing a bipartite graph of accident cases and evaluation indexes based on the accident case data and the evaluation indexes of the chemical enterprises;
calculating the correlation strength of the accident node and the evaluation index by using a recommendation algorithm based on matrix decomposition based on the bipartite graph of the accident case and the evaluation index; calculating the influence of the accident node;
processing the bipartite graph by adopting a PersonalRank algorithm of random walk based on the correlation strength between the accident node and the historical evaluation indexes and the influence of the accident node to obtain an optimized weight of each evaluation index;
carrying out quantitative evaluation on the chemical enterprises by adopting the optimized evaluation indexes, and calculating the accident risk level of the chemical enterprises to be evaluated according to the evaluation score of each index to obtain the potential safety hazards in the safety production of the chemical enterprises; and sending the accident risk level and the potential safety hazard of the chemical enterprise to be evaluated to the mobile terminal of the corresponding manager.
In a second aspect, the invention provides a chemical enterprise accident risk assessment system based on index weight optimization;
chemical industry enterprise accident risk assessment system based on index weight optimization includes:
a build module configured to: constructing a bipartite graph of the accident cases and the evaluation indexes based on the accident case data and the evaluation indexes of the chemical enterprises;
an association strength calculation module configured to: calculating the correlation strength of the accident node and the evaluation index by using a recommendation algorithm based on matrix decomposition based on the bipartite graph of the accident case and the evaluation index; calculating the influence of the accident node;
a weight optimization module configured to: processing the bipartite graph by adopting a PersonalRank algorithm of random walk based on the correlation strength between the accident node and the historical evaluation indexes and the influence of the accident node to obtain an optimized weight of each evaluation index;
a risk assessment module configured to: carrying out quantitative evaluation on the chemical enterprises by adopting the optimized evaluation indexes, and calculating the accident risk level of the chemical enterprises to be evaluated according to the evaluation score of each index to obtain the potential safety hazards in the safety production of the chemical enterprises; and sending the accident risk level and the potential safety hazard of the chemical enterprise to be evaluated to the mobile terminal of the corresponding manager.
In a third aspect, the present invention further provides an electronic device, including: one or more processors, one or more memories, and one or more computer programs; wherein a processor is connected to the memory, the one or more computer programs are stored in the memory, and when the electronic device is running, the processor executes the one or more computer programs stored in the memory, so as to make the electronic device execute the method according to the first aspect.
In a fourth aspect, the present invention also provides a computer-readable storage medium for storing computer instructions which, when executed by a processor, perform the method of the first aspect.
Compared with the prior art, the invention has the beneficial effects that:
the method is developed and researched aiming at the problem of optimal setting of index weight in a safety assessment system of the dangerous chemical accident, and provides a weight adjusting method based on bipartite graphs associated with accident cases and evaluation indexes, so that important hidden dangers existing in production management of chemical enterprises can be accurately fed back. Thereby being capable of essentially reducing the occurrence of accidents in chemical production. And secondly, forming an accident sample-index system association score matrix by evolving the association bipartite graph, and supplementing and perfecting values in the association score matrix by using a recommendation algorithm based on matrix decomposition so as to obtain the association strength between complete evaluation index nodes and accident nodes. Finally, the invention also designs an evaluation index weight increment optimization system, and the modeling method is integrated into the system, so that the increment optimization work of the evaluation index system weight can be completed more intuitively and conveniently, and the rapidness and traceability of the evaluation index system weight optimization work are effectively improved.
Safety management plays a vital role in enterprises, a perfect enterprise safety management system is constructed, the guiding thought, the target and the behavior of a safety idea are regulated and solidified by utilizing the forms of rules and systems, the process of making the safety rules and systems of the enterprises is the process of well performing safety management, the safety management of the process is well performed, the safety management system needs to be continuously perfected, and whether the safety evaluation system is reasonable or not mainly depends on two elements: evaluation index and index weight. The evaluation indexes are scoring elements in an evaluation system, are potential safety management hazards possibly existing in the evaluation production process, and are incentive sets for generating accident risks. The index weight reflects the importance of the evaluation index in the occurrence of the accident risk. The greater the weight of the index, the greater the likelihood that an accident will be caused by the index. Therefore, it is especially important to continuously optimize and modify each index and its weight value of the safety management index system, and continuously optimizing and dynamically adjusting the index weight value can improve the quality and efficiency of safety management work, so that the enterprise safety management work moves toward standardization, systematization and programming.
Aiming at the problem of weight optimization of an index system for safety quantitative evaluation in the chemical industry, the invention calculates the influence of accident nodes by integrating the occurrence time of accident case samples and the hazard degree of accidents. The method comprises the steps of constructing a network structure model of an accident and index system through correlation analysis of accident root reasons determined by accident analysts and index items related to the index system, constructing and forming an accident sample-index system correlation scoring matrix through evolution accident cases and evaluation index bipartite graphs, and completing values of the scoring matrix by adopting a recommendation algorithm based on matrix decomposition for calculating the correlation strength between evaluation index nodes and accident nodes in consideration of the phenomena of accident sample-index system correlation scoring omission or missing due to the fact that the analysts are different in individuals and incomplete in analysis. And designing an index weight increment optimization algorithm based on Personalrank random walk, and dynamically adjusting and optimizing the weight of the index item under the weight constraint of an index system. The accuracy of the evaluation weight value adjustment is ensured, and the safety quantitative evaluation system can objectively and truly evaluate the production management and risk prevention level.
The invention measures the influence of all accidents on the index item set when calculating the index weight optimization. Considering that the traditional PersonalRank calculates the influence of a single accident, and meanwhile, because the traditional PersonalRank algorithm does not consider the initial weight of the node and does not distinguish the weight of the connecting edge, the invention redesigns the PersonalRank algorithm by considering the factors on the basis of the traditional PersonalRank algorithm, constructs the associated bipartite graph of the accident case and the evaluation index, forms an accident sample-index system associated score matrix by evolving the associated bipartite graph, and completes the value in the associated score matrix by using a recommendation algorithm based on matrix decomposition so as to obtain the complete association strength between the evaluation index node and the accident node. Calculating the influence of the accident node by integrating the occurrence time of the accident case sample and the hazard degree of the accident; finally, an index weight optimization algorithm based on the PersonalRank is formed. Finally, the invention also designs an evaluation index weight increment optimization system, and the modeling method is integrated into the system, so that the increment optimization work of the evaluation index system weight can be completed more intuitively and conveniently, and the rapidness and the traceability of the evaluation index system weight optimization work are effectively improved.
The method analyzes the index system and the weight of the index items thereof according to policy and actual management requirements, and performs dynamic fine adjustment or correction, thereby ensuring the scientific, reasonable and feasible safety risk assessment system of the chemical enterprises, promoting the continuous improvement of enterprise supervision and further better avoiding risks.
The invention provides an index standard weight optimization method and system for chemical enterprise accident risk assessment based on objective facts of root cause analysis of a large number of accidents for a period of time. Based on the invention, the method is also helpful for discovering the potential safety hazard and risk of the whole safety production in the chemical industry, and is helpful for assisting the safety supervision department to make reasonable supervision measures and perfect the safety law system.
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.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
FIG. 1 is a flow chart of the method and system for optimizing the quantitative evaluation index weight increment for safety of chemical enterprises based on accident correlation;
FIG. 2 is a bipartite graph relating chemical accident cases and evaluation indexes.
Detailed Description
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, 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 belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular is intended to include the plural unless the context clearly dictates otherwise, and furthermore, it should be understood that the terms "comprises" and "comprising," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
The embodiments and features of the embodiments of the present invention may be combined with each other without conflict.
Example one
The embodiment provides a chemical enterprise accident risk assessment method based on index weight optimization;
as shown in fig. 1, the method for evaluating the accident risk of the chemical industry enterprise based on index weight optimization includes:
s101: constructing a bipartite graph of the accident cases and the evaluation indexes based on the accident case data and the evaluation indexes of the chemical enterprises;
s102: calculating the correlation strength of the accident node and the evaluation index by using a recommendation algorithm based on matrix decomposition based on the bipartite graph of the accident case and the evaluation index; calculating the influence of the accident node;
s103: based on the correlation strength between the accident node and the historical evaluation index and the influence of the accident node, processing the bipartite graph by adopting a random walk PersonalRank algorithm to obtain an optimized weight of each evaluation index;
s104: carrying out quantitative evaluation on the chemical enterprises by adopting the optimized evaluation indexes, and calculating the accident risk level of the chemical enterprises to be evaluated according to the evaluation score of each index to obtain the potential safety hazards in the safety production of the chemical enterprises; and sending the accident risk level and the potential safety hazard of the chemical enterprise to be evaluated to the mobile terminal of the corresponding manager.
Further, the method further comprises:
s105: acquiring new accident case data of a chemical enterprise, wherein the accident case data comprises the following steps: death people, injury people, property loss amount, accident cause and accident type; and repeating S101 to S104, and updating the evaluation index weight in an iterative manner.
Further, the S101: constructing a bipartite graph of accident cases and evaluation indexes based on the accident case data and the evaluation indexes of the chemical enterprises; the method specifically comprises the following steps:
constructing a bipartite graph between the accident case and the evaluation index through the incidence relation between the accident root cause and the evaluation index item of the safety evaluation system;
in the bipartite graph construction process, both the accident root cause and the evaluation index item are regarded as nodes, if the incidence relation exists between the accident root cause and the evaluation index item, a connecting line exists between the two nodes, and if the incidence relation does not exist between the accident root cause and the evaluation index item, the connecting line does not exist between the two nodes.
Illustratively, through a large amount of investigation on accident case induction causes, the incidence relation between each accident case and the safety assessment system index item is determined, and a bipartite graph for incidence case and evaluation index correlation is formed.
By D ═ D 1 ,d 2 ,…,d m Denotes a set of m accident samples, where d i Is the ith accident case sample; l ═ L 1 ,l 2 ,…,l n Is a set of n index items of an evaluation system, wherein l j Is the jth evaluation index item.
The accident case and the evaluation index are associated with a bipartite graph which is expressed as follows:
let DL _ BiGraph be (V, E) called incident case and evaluation index bipartite graph, V being a set of nodes, E being a set of edges, if and only if:
(1)V＝V D ∪V L in which V is D For accident case node binding, V L Is a set of evaluation index nodes, and V D ∩V L ＝Φ；
(2)
If v is i ∈V D Then v is j ∈V L 。
Further, the S102: calculating the correlation strength of the accident node and the evaluation index by using a recommendation algorithm based on matrix decomposition based on the bipartite graph of the accident case and the evaluation index; the method specifically comprises the following steps:
setting and optimizing the weight of two types of nodes of accident root reasons and evaluation index items in the bipartite graph, and obtaining an accident sample and index system association score sparse matrix according to the bipartite graph;
and supplementing and perfecting the values in the association score sparse matrix by using a recommendation algorithm based on matrix decomposition, so as to obtain the association strength between the accident node and the evaluation index.
Further, the S102: calculating the correlation strength of the accident node and the evaluation index by using a recommendation algorithm based on matrix decomposition based on the bipartite graph of the accident case and the evaluation index; the detailed steps comprise:
s1021: the given chemical accident risk assessment system is provided with n evaluation indexes, and the evaluation indexes are expressed as I ═ I 1 ,I 2 ,...,I n Let the accident case set D ═ D 1 ,D 2 ,...,D m M is the number of accident cases;
s1022: the expert personnel set with the evaluation index weight is set as U ═ U 1 ,u 2 ,...,u s S is the number of experts participating in the evaluation;
s1023: constructing an accident case and evaluation index bipartite graph, and evolving the bipartite graph to form an accident sample-index system association score sparse matrix;
s1024: and constructing an incidence relation between the nodes in the accident case set D and the index item set L, grading by the expert U, and supplementing and perfecting values in an incidence grading sparse matrix by using a recommendation algorithm based on matrix decomposition so as to obtain the incidence strength between the accident nodes and the evaluation indexes.
Because the accident case and the evaluation index matrix are huge, experts cannot accurately score all the evaluation indexes related to each accident. Design expert u k Part of accidents D according to experience i Corresponding evaluation index I j Giving partial expert scoring values; the method includes the steps that a matrix decomposition algorithm is utilized to supplement and perfect scores of sparse correlation score matrixes of an accident sample-index system with relatively incomplete scores of experts, and therefore correlation strength values between complete evaluation index nodes and accident nodes are obtained.
Further, the S1024: constructing an incidence relation between nodes in the accident case set D and the index item set L, grading by an expert U, and calculating the incidence strength between the accident nodes and the evaluation indexes by using a recommendation algorithm based on matrix decomposition; the method comprises the following specific steps:
s10241: the system comprises m accident cases and n evaluation indexes, wherein the n evaluation indexes form an m x n accident sample-index system correlation scoring matrix, accident nodes are rows of the matrix, and index nodes are columns of the matrix;
s10242: and (4) setting the correlation score set of the expert on the accident sample-index system to be {5, 4, 3, 2, 1}, wherein the higher the score is, the more the expert considers that the index has greater influence on the accident.
S10243: let expert u k Part of the accidents D according to experience i Corresponding evaluation index I j Giving out expert scoring values to form an accident sample-index system association scoring sparse matrix R Sparse ：
R Sparse ＝(r ij ) m×n (1)。
S10244: using a recommendation algorithm based on matrix decomposition to perform m x n dimensional matrix R Sparsity And decomposing the data into two low-dimensional matrixes m x f and f x n, wherein m is the accident number, n is the evaluation index number, and f represents the hidden vector dimension.
Accident cases andthe evaluation indicators are all mapped to a joint latent factor space of dimension f so that accident cases and evaluation indicators are interactively modeled as internal products in the joint latent factor space. Thus, each accident case d i And a vector q i Associated with each evaluation index l j And a vector p j Is associated q i And p j Are real vectors in the f dimension.
For a given ith accident case, q i As an accident d i Embedding the vector representation of f-dimensional implicit characteristic space of the interaction between the accident and the evaluation index;
for a given evaluation index j, p j As an evaluation index l j Vector representation of an f-dimensional implicit feature space embedded into the interaction between the accident and the evaluation index;
s10245: computing
Dot product of (1), obtained
The dot product of (1) captures the evaluation index l j And accident case d i Influence degree therebetween, i.e. evaluation index l j For case d of induced accident i The importance of (c).
After a chemical accident occurs, an evaluation index l is set j For accident case d i Predictive scoring of degree of influence of (3)
(define true value as r) ij ) So as to obtain the predicted value of each evaluation index j:
s10246: calculating prediction score by using mean square error and adding regularization item
With a true score r ij Loss function of (2):
wherein λ is regularization hyper-parameter, avoiding disturbance generated by extreme samples, and K is a set of all expert scoring samples.
S10247: solving q by gradient descent method i And p j 。
Setting a function:
for q in the loss function i And p j The partial derivatives are respectively calculated to obtain:
respectively substituting the two partial derivatives into gradient descent parameter updating formula
Obtaining:
in the above-mentioned formula, the first and second,
i.e. the error gamma of the current truth score and the prediction score i,j Eta is the learning rate, thereforeThe method is simple and available:
q i ＝q i +η(γ ij p j -λq i ) (9)
p j ＝p j +η(γ ij q i -λp j ) (10)
s10248: filling the accident sample-index system sparse association scoring matrix by using a recommendation algorithm based on matrix decomposition to obtain a complete accident sample-index system scoring matrix;
obtaining the correlation strength between complete evaluation index nodes and accident nodes according to the value of each element in the complete accident sample-index system scoring matrix;
wherein the content of the first and second substances,
for the prediction score, m is the number of accidents and n is the number of evaluation indicators.
Further, S102: the influence of the accident node is calculated; the method specifically comprises the following steps:
and (4) integrating the occurrence time of the accident case sample and the hazard degree of the accident, and calculating the influence of the accident node.
The calculation of the influence weight of the accident node of the bipartite graph related to the accident case and the evaluation index is obtained by synthesizing evaluation factors such as the occurrence time of the accident, the hazard degree of the accident and the like of the characteristic attributes of the accident.
Further, S102: the influence of the accident node is calculated; the detailed steps comprise:
synthesizing the occurrence time of the accident case sample and the hazard degree of the accident, and calculating the influence h of the accident node i ：
h i ＝T(t i )×s i (12)
Wherein s is i Quantitative value of accident hazard for the ith sample, T (T) i ) Time weights for accident case samplesThe accident case sample weight closer to the current time is larger, and the calculation formula is as follows:
T(t i )＝e -(ct-ti) (13)
wherein, t i Represents the occurrence time of the ith sample, and ct represents the current time.
Further, the step S103: based on the correlation strength between the accident node and the historical evaluation index and the influence of the accident node, processing the bipartite graph by adopting a random walk PersonalRank algorithm to obtain an optimized weight of each evaluation index; the method specifically comprises the following steps:
suppose bipartite graph G ═ V D ,V L ,E)，V D Representing a set of accident nodes, V L Representing index item set, E representing correlation between accident node and index item, and influence Q of accident node D ＝{h i I is not less than 1 and not more than M and an edge weight matrix Z is not less than Z i,j |1≤i≤M and 1≤j≤N}；
Define Γ as a weighting function, i.e.:
For the initial state of the process, the initial state,
the initial value is 0, i.e., Γ (v) j ) Each index item node is initialized to 0;
selecting bipartite graph node v i Starting to walk, when walking to a node, firstly deciding whether to continue walking according to the probability phi or stop the walking and go from v i The node restarts to walk;
if the next wandering is decided, randomly selecting one node from the nodes pointed by the current node as the next wandering node according to uniform distribution;
after a number of random walks, each v j The probability convergence of the node being visited tends to be stable;
based on the correlation strength matrix between the evaluation index node and the accident node, the elements in the matrix are weight calculation factors z ij The optimized Personalrank algorithm has the following calculation mode:
through multiple rounds of random walk of Personalrank, each index node v of the index item set j All obtain v i Degree of influence thereon
x j Represents the access probability of node j, phi is the probability of random walk, | out k L represents the sum of all outgoing edges of node k, and in (j) represents the set of incoming edges pointing to node j.
When the influence of all samples on the index items is calculated, the occurrence time of the accident case samples and the hazard degree of the accident are comprehensively considered, namely the influence of each accident node, and all M accident samples on the index l j The node impact of the weights is:
based on the dynamic adjustment algorithm of the original index system, the weight ratio pw of the index item j of M accidents is synthesized j The calculation method is as follows:
wherein the content of the first and second substances,
representing the proportion of the jth evaluation index in the original index system, and setting M 0 The set primary index system depends on the sample number parameter, T (T) 0 ) Is the time parameter of the original index system.
Calculating the optimized weight w of each index item of the index system j ：
w j ＝pw j ×SumW (18)
Wherein SumW is the sum of the weight of the index items of the constrained index system.
Further, the S104: carrying out quantitative evaluation on the chemical enterprises by adopting the optimized evaluation indexes, and calculating the accident risk level of the chemical enterprises to be evaluated according to the evaluation score of each index; the method specifically comprises the following steps:
and weighting and summing the evaluation quantitative results of the chemical enterprises to be evaluated according to the optimized weight of each evaluation index, and matching the summed result with the preset accident risk level to obtain the accident risk level of the chemical enterprises to be evaluated.
The evaluation index in the safety evaluation system can objectively and accurately reflect the potential hazards possibly existing in the safety production and management process of the enterprise. The weight of the evaluation index reflects the importance of the evaluation index when the potential safety hazard occurs. In the daily production management of enterprises, safety accidents or hidden dangers are inevitable. After a safety accident occurs, the accident that has occurred and the potential safety hazard that has occurred are generally analyzed and evaluated according to a safety evaluation system, and an accident case report is formed. In the accident case report, the safety analyst associates the root cause of the safety accident with the safety evaluation index, and evaluates the importance of the safety index of the accident. Therefore, the incidence relations between the accident cases and the evaluation indexes are mined from the accident case reports, the incidence relations are reasonably and quantitatively evaluated, and the weights of the evaluation indexes are optimized and adjusted based on the information. Through a large amount of investigation of accident case induction root causes, and analysis of various factors such as accident types, accident cause mechanisms, accident hazards and the like, the association relationship between each accident case and the index items of the safety assessment system can be determined, and a bipartite graph of the association between the accident cases and the evaluation indexes shown in fig. 2 is formed.
In the bipartite graph, node types are divided into accident case nodes and evaluation index nodes, and a connection arc between the two types of nodes represents the corresponding relation between the accident case induction root cause and the evaluation index items. E.g. accident case node d 3 And evaluation index node l 1 ，l 4 And l n If correlated, then represent d 3 The root cause of the accident includes an evaluation index node l 1 ，l 4 And l n The corresponding content.
The accident case sample comprises the occurrence time of the accident sample, the type of the related enterprise, the accident cause mechanism, the accident hazard and the like. In the accident case and evaluation index bipartite graph, two concepts of node influence and correlation strength are defined for accident nodes.
The accident node influence is used for depicting the contribution degree of the accident case corresponding to the node to the index weight correction. The strength of association between the accident case node and the evaluation index node refers to the closeness degree between the evaluation index node and the accident node, and the greater the strength of association, the higher the contribution degree of the evaluation index node to the accident corresponding to the induction node is.
In the present invention, the degree of contribution of the accident case to the index weight correction is determined by the accident occurrence time, the risk, and the reliability of the accident analysis. The correlation strength is determined by evaluating the correlation importance of the accident root cause and the index item. A calculation method of node influence and correlation strength is a main basis for weight optimization and adjustment by using a PersonalRank random walk algorithm.
The accident analysis personnel judges the root cause of each accident case and the related importance of the index item of the evaluation system, when the importance is judged, the accident sample-index system related score set {5, 4, 3, 2, 1} is utilized, and different experts can select different scores to carry out related judgment on the importance of the index according to professional knowledge and practical experience of the experts. The higher the score, the greater the influence of the index on the occurrence of the accident, which the representative expert considers.
The incidence bipartite graph is evolved to form an accident sample-index system incidence score sparse matrix, and values in the sparse incidence score matrix are completed by utilizing a recommendation algorithm based on matrix decomposition in consideration of the individual difference and incomplete analysis of analysts and the phenomena of accident sample-index system incidence score omission or loss, so that the correlation strength between complete evaluation index nodes and accident nodes is obtained.
The accident node influence is calculated by integrating the occurrence time of the accident case sample and the hazard degree of the accident, and the problems of inconsistent importance comprehensive evaluation credibility and weight proportion of the characteristic attributes of the accident case sample are solved, wherein the problems are given by the uneven knowledge level and working experience level of accident analysts.
The relation between the chemical safety management index system and the accident case is a bipartite graph model with weight, and the quantified accident sample node weight and the accident sample associated index item value
And (3) comprehensively considering factors of credibility, time and hazard degree of chemical accident sample nodes by using a binary graph-based random walk PersonalRank algorithm, and then forming an index weight optimization method based on PersonalRank.
For example, in the embodiment, the optimized index weight determined by the random walk method, the original weight obtained by calculation and the optimized weight table after the random walk-based calculation are as follows:
TABLE 1 original weight and weight table optimized based on random walk calculation
In order to carry out the safety management system of an enterprise in the whole process of safe production and operation of the enterprise, the safety management work of the enterprise is better implemented and developed, and the safety work of the enterprise is better prevented from happening, the index weight is optimized by using the algorithm of the invention, and further the suggested modification value of the index weight, namely the optimized weight of an accident case, is obtained, taking 540 three-level index items of an A safety institute as an example, and the modification suggestion obtained by the optimization algorithm is as follows: the index items needing to improve the original weight are 111 items, the index items needing to reduce the original weight are 126 items, and the index items needing to be unchanged in weight are 303 items.
And analyzing a modification suggestion score obtained by optimizing the algorithm, analyzing the content included in hidden danger investigation of a secondary index item 2.7 ' hidden danger investigation and treatment ' of a tertiary index item 2.7.1 ' hidden danger investigation ' of a primary index item 2 ' risk assessment ' of an original index system provided by the A's institute of safety, wherein the original weight value of the index item is 44, and the value after the optimization of the index weight optimization algorithm is 95.
Example two
The embodiment provides a chemical enterprise accident risk assessment system based on index weight optimization;
chemical enterprise accident risk assessment system based on index weight optimization includes:
a build module configured to: constructing a bipartite graph of accident cases and evaluation indexes based on the accident case data and the evaluation indexes of the chemical enterprises;
an association strength calculation module configured to: calculating the correlation strength of the accident node and the evaluation index by using a recommendation algorithm based on matrix decomposition based on the bipartite graph of the accident case and the evaluation index; calculating the influence of the accident node;
a weight optimization module configured to: processing the bipartite graph by adopting a PersonalRank algorithm of random walk based on the correlation strength between the accident node and the historical evaluation indexes and the influence of the accident node to obtain an optimized weight of each evaluation index;
a risk assessment module configured to: carrying out quantitative evaluation on the chemical enterprises by adopting the optimized evaluation indexes, and calculating the accident risk level of the chemical enterprises to be evaluated according to the evaluation score of each index to obtain the potential safety hazards in the safety production of the chemical enterprises; and sending the accident risk level and the potential safety hazard of the chemical enterprise to be evaluated to the mobile terminal of the corresponding manager.
It should be noted here that the building module, the association strength calculating module, the quantifying module, the weight optimizing module and the risk evaluating module correspond to steps S101 to S104 in the first embodiment, and the modules are the same as the corresponding steps in the implementation example and the application scenario, but are not limited to the disclosure in the first embodiment. It should be noted that the modules described above as part of a system may be implemented in a computer system such as a set of computer-executable instructions.
In the foregoing embodiments, the description of each embodiment has an emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions in other embodiments.
The proposed system can be implemented in other ways. For example, the above-described system embodiments are merely illustrative, and for example, the division of the above-described modules is merely a logical division, and in actual implementation, there may be other divisions, for example, multiple modules may be combined or integrated into another system, or some features may be omitted, or not executed.
EXAMPLE III
The present embodiment also provides an electronic device, including: one or more processors, one or more memories, and one or more computer programs; wherein, a processor is connected with the memory, the one or more computer programs are stored in the memory, and when the electronic device runs, the processor executes the one or more computer programs stored in the memory, so as to make the electronic device execute the method according to the first embodiment.
It should be understood that in this embodiment, the processor may be a central processing unit CPU, and the processor may also be other general purpose processor, a digital signal processor DSP, an application specific integrated circuit ASIC, an off-the-shelf programmable gate array FPGA or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory may include both read-only memory and random access memory, and may provide instructions and data to the processor, and a portion of the memory may also include non-volatile random access memory. For example, the memory may also store device type information.
In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in a processor or instructions in the form of software.
The method in the first embodiment may be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules in the processor. The software modules may be located in ram, flash, rom, prom, or eprom, registers, among other storage media as is well known in the art. The storage medium is located in a memory, and a processor reads information in the memory and completes the steps of the method in combination with hardware of the processor. To avoid repetition, it is not described in detail here.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
Example four
The present embodiments also provide a computer-readable storage medium for storing computer instructions, which when executed by a processor, perform the method of the first embodiment.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (9)
1. The chemical enterprise accident risk assessment method based on index weight optimization is characterized by comprising the following steps:
constructing a bipartite graph of accident cases and evaluation indexes based on the accident case data and the evaluation indexes of the chemical enterprises;
calculating the correlation strength of the accident case and the evaluation index by using a recommendation algorithm based on matrix decomposition based on the bipartite graph of the accident case and the evaluation index; calculating the influence of the accident case;
processing the bipartite graph by adopting a PersonalRank algorithm of random walk based on the correlation strength between the accident case and the evaluation indexes and the influence of the accident case to obtain an optimized weight of each evaluation index;
carrying out quantitative evaluation on the chemical enterprises by adopting the optimized evaluation indexes, and calculating the accident risk level of the chemical enterprises to be evaluated according to the evaluation score of each index to obtain the potential safety hazards in the safety production of the chemical enterprises; sending the accident risk level and the potential safety hazard of the chemical enterprise to be evaluated to the mobile terminal of the corresponding manager;
processing the bipartite graph by adopting a PersonalRank algorithm of random walk based on the correlation strength between the accident case and the evaluation indexes and the influence of the accident case to obtain an optimized weight of each evaluation index; the method specifically comprises the following steps:
assuming that the bipartite graph G is (D, I, E), D represents an accident case set, I represents an index item set, E represents the association of the accident case and the index item, and the influence Q of the accident case D ＝{h i I is not less than 1 and not more than m and an edge weight matrix Z is not less than Z i,j |1≤i≤m∧1≤j≤n}；
Define Γ as a weight function:
For the initial state of the process, the initial state,
the initial value is 0, Γ (I) j ) Index term I is 0 j The initial value of the weight is 0;
selecting bipartite graph node D i Starting to walk, when walking to a node, firstly according to the probability
Deciding whether to continue the walk, or to stop the walk and go from D i The node restarts to walk;
if the next wandering is decided, randomly selecting one node from the nodes pointed by the current node as the next wandering node according to uniform distribution;
after multiple random walks, the probability convergence of each node being accessed tends to be stable;
based on the correlation strength matrix between the evaluation index items and the accident cases, the elements in the matrix are weight calculation factors z ij (ii) a The optimized Personalrank algorithm has the following calculation mode:
when C is present j When the index is an index item, the value is 1; when C is present j In case of accident, the value is 0;
after multiple rounds of random walk of Personalrank, each index item I of the index item set j All obtain v i Degree of influence thereon
x k Which represents the probability of access for node k,
is the probability of random walk, | out k L represents the sum of all outgoing edges of the node k, and in (j) represents an incoming edge set pointing to the node j;
when the influence of all samples on the index items is calculated, the occurrence time of the accident case samples and the hazard degree of the accident are comprehensively considered;
index items I of m accidents are synthesized based on dynamic adjustment algorithm of original index system j Weight ratio pw of j ；
Wherein the content of the first and second substances,
representing the proportion of the jth evaluation index in the original index system, and setting M 0 For a set primitive index system, T (T) is a sample number parameter 0 ) Time parameter of original index system;
calculating the optimized weight w of each index item of the index system j ：
w j ＝pw j ×SumW (5)
Wherein, SumW is the sum of the index item weights of the constrained index system.
2. The chemical industry enterprise accident risk assessment method based on index weight optimization of claim 1, wherein a bipartite graph of accident cases and evaluation indexes is constructed based on accident case data and evaluation indexes of chemical industry enterprises; the method specifically comprises the following steps:
constructing a bipartite graph between an accident case and an evaluation index through an incidence relation between the accident root cause and the evaluation index item of a security evaluation system;
in the bipartite graph construction process, both the accident root cause and the evaluation index item are regarded as nodes, if the incidence relation exists between the accident root cause and the evaluation index item, a connecting line exists between the two nodes, and if the incidence relation does not exist between the accident root cause and the evaluation index item, the connecting line does not exist between the two nodes.
3. The chemical industry enterprise accident risk assessment method based on index weight optimization of claim 1, wherein based on the bipartite graph of the accident cases and the evaluation indexes, the correlation strength of the accident cases and the evaluation indexes is calculated by using a recommendation algorithm based on matrix decomposition; the method specifically comprises the following steps:
setting and optimizing weights of two types of nodes of accident root reasons and evaluation index items in the bipartite graph, and obtaining an accident case and index system association score sparse matrix according to the bipartite graph; and supplementing and perfecting the values in the association score sparse matrix by using a recommendation algorithm based on matrix decomposition, so as to obtain the association strength between the accident case and the evaluation index.
4. The chemical industry enterprise accident risk assessment method based on index weight optimization according to claim 1, wherein based on the bipartite graph of the accident cases and the evaluation indexes, the correlation strength of the accident cases and the evaluation indexes is calculated by using a recommendation algorithm based on matrix decomposition; the detailed steps comprise:
the given chemical accident risk assessment system is provided with n evaluation indexes, and the evaluation indexes are expressed as I ═ I 1 ,I 2 ,...,I n Let D ═ D for accident case set 1 ,D 2 ,...,D m M is the number of accident cases;
the expert personnel set with the evaluation index weight is U ═ U 1 ,u 2 ,...,u s S is the number of experts participating in the evaluation;
constructing an accident case and evaluation index bipartite graph, and evolving the bipartite graph to form an accident case-index system association score sparse matrix;
and (3) constructing an incidence relation between nodes in the accident case set D and the index item set I, scoring by the expert U, and supplementing and perfecting values in an incidence scoring sparse matrix by using a matrix decomposition algorithm so as to obtain the incidence strength of the accident case and the evaluation index.
5. The chemical enterprise accident risk assessment method based on index weight optimization as claimed in claim 1, wherein an incidence relation between nodes in the accident case set D and the index item set I is constructed, an expert U scores, and a recommendation algorithm based on matrix decomposition is used for supplementing and perfecting values in an incidence score sparse matrix, so as to obtain the incidence strength between the accident cases and the evaluation indexes; the method comprises the following specific steps:
the system comprises m accident cases and n evaluation indexes, wherein the n evaluation indexes form an m x n accident case-index system correlation scoring matrix, the accident cases are rows of the matrix, and the index items are columns of the matrix;
the correlation evaluation set of the expert on the accident case-index system is set as {5, 4, 3, 2, 1}, wherein the higher the score is, the larger the influence of the index on the accident is considered by the expert;
let expert u k Experience-based partial accident case D i Corresponding evaluation index I j Giving expert rating values to form an accident case-index system association rating sparse matrix R Sparsity ；
Using matrix decomposition algorithm to divide the matrix R into m x n dimensions Sparse Decomposing the data into two low-dimensional matrixes m x f and f x n, wherein m is the number of accident cases, n is the number of evaluation indexes, and f represents the dimension of a hidden vector;
the accident case and the evaluation index are mapped to a joint potential factor space with the dimension f so as to interactively model the accident case and the evaluation index into an internal product in the joint potential factor space; thus, each accident case D i And a vector q i In the association of the data packets with each other,each evaluation index I j And a vector p j Are correlated, q i And p j Real number vectors that are all f-dimensional;
for a given ith accident case, q i For accident case D i Embedding the vector representation of the f-dimensional joint potential factor space of the accident and evaluation index interaction;
for a given j-th evaluation index, p j As an evaluation index I j Vector representation of an f-dimensional joint potential factor space embedded into the interaction of the accident and the evaluation index;
computing
Dot product of (1), obtained
The dot product of (1) captures the evaluation index I j And accident case D i Degree of influence therebetween, evaluation index I j For induced accident case D i The importance of (c);
calculating prediction score by using mean square error and adding regularization item
With true score r ij A loss function of (d);
solving q by gradient descent method i And p j ；
Filling the accident case-index system sparse association scoring matrix by using a recommendation algorithm based on matrix decomposition to obtain a complete accident case-index system scoring matrix;
and obtaining the correlation strength between the complete evaluation index item and the accident case according to the value of each element in the complete accident case-index system scoring matrix.
6. The chemical industry enterprise accident risk assessment method based on index weight optimization of claim 1, wherein the influence of accident case is calculated; the method specifically comprises the following steps:
integrating the occurrence time of the accident case sample and the hazard degree of the accident, and calculating the influence of the accident case;
the calculation of the influence weight of the accident case of the bipartite graph associating the accident case with the evaluation index is obtained by synthesizing the characteristic attributes of the accident itself, including the occurrence time of the accident and the risk evaluation factor of the accident.
7. The chemical enterprise accident risk assessment system based on index weight optimization, which adopts the chemical enterprise accident risk assessment method based on index weight optimization according to claim 1, is characterized by comprising:
a build module configured to: constructing a bipartite graph of accident cases and evaluation indexes based on the accident case data and the evaluation indexes of the chemical enterprises;
an association strength calculation module configured to: calculating the correlation strength of the accident case and the evaluation index by using a recommendation algorithm based on matrix decomposition based on the bipartite graph of the accident case and the evaluation index; calculating the influence of the accident case;
a weight optimization module configured to: based on the correlation strength between the accident case and the evaluation indexes and the influence of the accident case, processing the quantized bipartite graph by adopting a Personalrank algorithm of random walk to obtain an optimized weight of each evaluation index;
a risk assessment module configured to: carrying out quantitative evaluation on the chemical enterprises by adopting the optimized evaluation indexes, and calculating the accident risk level of the chemical enterprises to be evaluated according to the evaluation score of each index to obtain the potential safety hazards in the safety production of the chemical enterprises; and sending the accident risk level and the potential safety hazard of the chemical enterprise to be evaluated to the mobile terminal of the corresponding manager.
8. An electronic device, comprising: one or more processors, one or more memories, and one or more computer programs; wherein a processor is connected to the memory, the one or more computer programs being stored in the memory, the processor executing the one or more computer programs stored in the memory when the electronic device is running, to cause the electronic device to perform the method of any of the preceding claims 1-6.
9. A computer-readable storage medium storing computer instructions which, when executed by a processor, perform the method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110366696.XA CN113065771B (en) | 2021-04-06 | 2021-04-06 | Chemical enterprise accident risk assessment method and system based on index weight optimization |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110366696.XA CN113065771B (en) | 2021-04-06 | 2021-04-06 | Chemical enterprise accident risk assessment method and system based on index weight optimization |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113065771A CN113065771A (en) | 2021-07-02 |
| CN113065771B true CN113065771B (en) | 2022-08-19 |
Family
ID=76565827
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110366696.XA Active CN113065771B (en) | 2021-04-06 | 2021-04-06 | Chemical enterprise accident risk assessment method and system based on index weight optimization |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113065771B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113742788A (en) * | 2021-09-01 | 2021-12-03 | 宁波大学科学技术学院 | Product entrusted production system and method based on trusted data supervision |
| CN114154817B (en) * | 2021-11-18 | 2022-07-19 | 南京大学 | Automated enterprise environment risk assessment method and device |
| CN114880537B (en) * | 2022-05-13 | 2023-06-30 | 深圳宏鹏数字供应链管理有限公司 | Enterprise risk assessment method, system and storage medium |
| CN117151478B (en) * | 2023-10-31 | 2024-01-26 | 南京南工应急科技有限公司 | Chemical enterprise risk early warning method and system based on convolutional neural network |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103116639A (en) * | 2013-02-20 | 2013-05-22 | 新浪网技术(中国)有限公司 | Item recommendation method and system based on user-item bipartite model |
| CN106485096A (en) * | 2016-10-20 | 2017-03-08 | 中南大学 | MiRNA Relationship To Environmental Factors Forecasting Methodology based on random two-way migration and multi-tag study |
| CN107967571A (en) * | 2017-12-18 | 2018-04-27 | 青岛科技大学 | Index system generation method based on entropy assessment |
| CN109242306A (en) * | 2018-09-04 | 2019-01-18 | 深圳市城市公共安全技术研究院有限公司 | Safety production risk assessment method and system based on multilevel gray correlation analysis |
| CN110322053A (en) * | 2019-06-13 | 2019-10-11 | 华中科技大学 | A kind of score in predicting method constructing local matrix based on figure random walk |
-
2021
- 2021-04-06 CN CN202110366696.XA patent/CN113065771B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103116639A (en) * | 2013-02-20 | 2013-05-22 | 新浪网技术(中国)有限公司 | Item recommendation method and system based on user-item bipartite model |
| CN106485096A (en) * | 2016-10-20 | 2017-03-08 | 中南大学 | MiRNA Relationship To Environmental Factors Forecasting Methodology based on random two-way migration and multi-tag study |
| CN107967571A (en) * | 2017-12-18 | 2018-04-27 | 青岛科技大学 | Index system generation method based on entropy assessment |
| CN109242306A (en) * | 2018-09-04 | 2019-01-18 | 深圳市城市公共安全技术研究院有限公司 | Safety production risk assessment method and system based on multilevel gray correlation analysis |
| CN110322053A (en) * | 2019-06-13 | 2019-10-11 | 华中科技大学 | A kind of score in predicting method constructing local matrix based on figure random walk |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113065771A (en) | 2021-07-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113065771B (en) | Chemical enterprise accident risk assessment method and system based on index weight optimization | |
| WO2023019986A1 (en) | Science and technology service quality evaluation method and device based on combination weighting and fuzzy grey clustering | |
| CN109587713B (en) | Network index prediction method and device based on ARIMA model and storage medium | |
| Huang et al. | An empirical analysis of data preprocessing for machine learning-based software cost estimation | |
| CN110516910B (en) | Insurance policy and insurance model training method and insurance risk assessment method based on big data | |
| US20170300546A1 (en) | Method and Apparatus for Data Processing in Data Modeling | |
| CN106372952A (en) | Objective and subjective weight determining multi-model compositional verification-based enterprise credit assessment method and system | |
| CN109360105A (en) | Product risks method for early warning, device, computer equipment and storage medium | |
| CN109102201B (en) | Power distribution network input-output benefit evaluation method for voltage division grades | |
| CN107563645A (en) | A kind of Financial Risk Analysis method based on big data | |
| CN104837184A (en) | Heterogeneous wireless network selection method based on interval triangular fuzzy number | |
| CN113408869A (en) | Power distribution network construction target risk assessment method | |
| CN112668822A (en) | Scientific and technological achievement transformation platform sharing system, method, storage medium and mobile phone APP | |
| CN111967796A (en) | Food enterprise credit scoring method and device and electronic equipment | |
| CN111986026A (en) | Personal credit business risk assessment method and device based on analytic hierarchy process | |
| CN113822499A (en) | Train spare part loss prediction method based on model fusion | |
| CN114764682B (en) | Rice safety risk assessment method based on multi-machine learning algorithm fusion | |
| CN117132383A (en) | Credit data processing method, device, equipment and readable storage medium | |
| Li et al. | A comparative analysis of predictive data mining techniques | |
| El-Sheikh et al. | Four imputation methods for handling missing values in the ARDL model: An application on Libyan FDI | |
| Svabova et al. | Prediction of default of small companies in the Slovak Republic | |
| Pavelková et al. | Passenger car sales projections: Measuring the accuracy of a sales forecasting model | |
| CN111832854A (en) | Maturity quantitative evaluation method and system for automobile research and development quality management system and readable medium | |
| CN114091908A (en) | Power distribution network comprehensive evaluation method, device and equipment considering multi-mode energy storage station | |
| CN113159634A (en) | Financial product management method and device and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |