CN116703158A - Risk assessment model construction method for imported food risk early warning - Google Patents
Risk assessment model construction method for imported food risk early warning Download PDFInfo
- Publication number
- CN116703158A CN116703158A CN202310669481.4A CN202310669481A CN116703158A CN 116703158 A CN116703158 A CN 116703158A CN 202310669481 A CN202310669481 A CN 202310669481A CN 116703158 A CN116703158 A CN 116703158A
- Authority
- CN
- China
- Prior art keywords
- risk
- level
- index
- reject ratio
- weight
- 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.)
- Pending
Links
- 235000013305 food Nutrition 0.000 title claims abstract description 48
- 238000012502 risk assessment Methods 0.000 title claims abstract description 30
- 238000010276 construction Methods 0.000 title claims abstract description 21
- 238000011156 evaluation Methods 0.000 claims abstract description 32
- 230000008676 import Effects 0.000 claims abstract description 19
- 238000010801 machine learning Methods 0.000 claims abstract description 7
- 238000005516 engineering process Methods 0.000 claims abstract description 4
- 238000007689 inspection Methods 0.000 claims abstract description 4
- 238000000034 method Methods 0.000 claims description 79
- 238000001514 detection method Methods 0.000 claims description 47
- 230000008569 process Effects 0.000 claims description 41
- 239000011159 matrix material Substances 0.000 claims description 23
- 239000013598 vector Substances 0.000 claims description 15
- 238000013210 evaluation model Methods 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000011895 specific detection Methods 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 6
- 238000012935 Averaging Methods 0.000 claims description 3
- 238000013135 deep learning Methods 0.000 claims description 3
- 238000013278 delphi method Methods 0.000 claims description 3
- 230000005484 gravity Effects 0.000 claims description 3
- 238000011835 investigation Methods 0.000 claims description 3
- 238000013077 scoring method Methods 0.000 claims description 3
- 238000007726 management method Methods 0.000 description 7
- 230000006378 damage Effects 0.000 description 2
- 235000021393 food security Nutrition 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 241000607142 Salmonella Species 0.000 description 1
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 229910052785 arsenic Inorganic materials 0.000 description 1
- RQNWIZPPADIBDY-UHFFFAOYSA-N arsenic atom Chemical compound [As] RQNWIZPPADIBDY-UHFFFAOYSA-N 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- CYDMQBQPVICBEU-UHFFFAOYSA-N chlorotetracycline Natural products C1=CC(Cl)=C2C(O)(C)C3CC4C(N(C)C)C(O)=C(C(N)=O)C(=O)C4(O)C(O)=C3C(=O)C2=C1O CYDMQBQPVICBEU-UHFFFAOYSA-N 0.000 description 1
- CYDMQBQPVICBEU-XRNKAMNCSA-N chlortetracycline Chemical compound C1=CC(Cl)=C2[C@](O)(C)[C@H]3C[C@H]4[C@H](N(C)C)C(O)=C(C(N)=O)C(=O)[C@@]4(O)C(O)=C3C(=O)C2=C1O CYDMQBQPVICBEU-XRNKAMNCSA-N 0.000 description 1
- 229960004475 chlortetracycline Drugs 0.000 description 1
- 235000019365 chlortetracycline Nutrition 0.000 description 1
- 235000013365 dairy product Nutrition 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 238000009533 lab test Methods 0.000 description 1
- 235000020191 long-life milk Nutrition 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012954 risk control Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/02—Computing arrangements based on specific mathematical models using fuzzy logic
-
- 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
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- 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)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Health & Medical Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Fuzzy Systems (AREA)
- Molecular Biology (AREA)
- Primary Health Care (AREA)
- Algebra (AREA)
- Artificial Intelligence (AREA)
- Computational Mathematics (AREA)
- Automation & Control Theory (AREA)
- Evolutionary Computation (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a risk assessment model construction method for import food risk early warning, which belongs to the technical field of risk assessment model construction, and comprises the following specific steps: based on big data technology and machine learning algorithm, through analyzing import food declaration data, qualified evaluation data, laboratory inspection data, risk warning notification information of related institutions/departments and social public opinion related safety risk information, a Hadoop/Spark architecture and an ETL tool are adopted to realize real-time or quasi-real-time acquisition of multi-source and multi-type data.
Description
Technical Field
The invention relates to the technical field of risk assessment model construction, in particular to a risk assessment model construction method for imported food risk early warning.
Background
The united states has established a relatively perfect product quality security risk monitoring system and operation mechanism based on a risk early warning theory. The Consumer Product Safety Committee (CPSC) is responsible for quality supervision of the whole consumer product field, collects product safety information through manufacturer reporting system, sampling detection and other ways, establishes and maintains a database for public inquiry and retrieval, and carries out risk assessment and processing based on abundant data, and the management means mainly comprise fine, media exposure, recall of problematic products, start of legal programs and the like. The European Union has established special systems such as non-food consumer product rapid early warning system (RAPEX), food and feed early warning system (RASFF), medical appliances and medicines, etc., by means of which the European Union member nations can share corresponding early warning information. Taking RAPEX as an example, the European Union reports the dangerous situation of non-food products every week, and measures such as 'authority issuing alarm, producer recall, sales prohibition, market withdrawal, informing consumers, product recall' and the like are adopted to eliminate the harm. In addition, the European Union regularly perfects risk analysis standards, methods and guidelines to improve the accuracy and reliability of notification systems.
At present, the risk early warning research and practice of China in the aspect of import food safety admission management are relatively few. The main risk information data sources mainly comprise foreign TBT/SPS information, foreign customs return information, foreign regional or national risk early warning information, social feedback information, defective product recall information, product injury monitoring information, third party quality safety detection reports, quality safety risk information and the like provided by various industry organizations, technical institutions, industrial network forums and the like, information mastering is limited, and the risk information gathering channels which are communicated with the regulatory departments and societies are lacked, and the risk information has certain hysteresis. In addition, the system security risk problem discovery mechanism still needs to be perfected, and at present, the technical means of discovering industrial quality problems in advance by means of large data means and intelligent analysis of security risk data mainly by means of supervision spot check, social report, foreign early warning information tracking and the like in discovery of imported food security risk problems in China has not been deeply applied.
Currently, in the aspect of import food security access management, security risk information mainly has two acquisition approaches: firstly, by obtaining some risk warning reports of foreign or international organizations, secondly, reporting after a large number of problems are found in a specific checking/law enforcement link by customs/market supervision departments in various places, and most of information is not collected, managed, analyzed and judged by adopting an artificial intelligence means, so that hysteresis of information processing is easily caused. In addition, many information works are concerned with all-class commodities, and the specificity of imported food safety management is not fully considered.
Therefore, the invention discloses a risk assessment model construction method for import food risk early warning.
Disclosure of Invention
The present invention has been made in view of the above and/or the problems existing in the existing risk assessment model construction method for import food risk early warning.
Therefore, the present invention aims to provide a risk assessment model construction method for import food risk early warning, which can solve the above-mentioned existing problems.
In order to solve the technical problems, according to one aspect of the present invention, the following technical solutions are provided:
a risk assessment model construction method for import food risk early warning comprises the following specific steps:
step one: based on big data technology and machine learning algorithm, through analyzing import food declaration data, qualified evaluation data, laboratory inspection data, risk warning notification information of related institutions/departments and social public opinion related safety risk information, a Hadoop/Spark architecture and an ETL tool are adopted to realize real-time or quasi-real-time acquisition of multi-source and multi-type data, and an HBase distributed database is adopted to store and retrieve mass multi-type data;
step two: the python is used as a basic tool for big data modeling, a scientific and effective imported food risk assessment model is built on the basis of a scikit-learn/tensorflow machine learning/deep learning framework, 2 different models are adopted for assessing risk on imported food, a comprehensive evaluation model with laboratory data is built on the basis of laboratory detection data, and a fuzzy comprehensive evaluation model without laboratory data is built on the basis of an AHP-risk matrix.
As a preferred scheme of the risk assessment model construction method for import food risk early warning, the invention is characterized in that: the concrete flow of the construction of the comprehensive evaluation model based on the laboratory detection data is as follows:
the process is as follows: comparing the laboratory detection result with a standard specified limit value, and respectively calculating to obtain the reject ratio and reject ratio of the third-level index, the second-level index and the first-level index;
a second flow: after the reject ratio and reject ratio of the three-level indexes are calculated, a weighting method is adopted to obtain the risk value R of each level of indexes, wherein the larger the risk value R is, the higher the risk is;
and a process III: the weight is calculated by adopting a Deltaffy-entropy method, wherein the weight is calculated according to the discrete degree of the index by means of the concept of entropy in the Deltaffy-entropy method, and the reject ratio weight calculated by using the Deltaffy-entropy method are respectively W X2 、W Y2 ;
The process is four: in order to avoid weight distortion of the Delphi-entropy method, a Delphi expert scoring method is combined, and the reject ratio and reject weight obtained by the Delphi method are respectively W X1 、W Y1 ;
The fifth flow is: calculating final weight W of reject ratio and reject ratio by using formula I X 、W Y ;
The sixth flow: after the reject ratio, reject ratio and weight value of each level of index are obtained, the risk value of each level of index is obtained by using a weighted summation mode, so that the risk level and the detection item are determined.
As a preferred scheme of the risk assessment model construction method for import food risk early warning, the invention is characterized in that: the reject ratio reflects the number of occurrences of reject, and the reject ratio reflects the degree to which the detected value deviates from the standard limit value.
As a preferred scheme of the risk assessment model construction method for import food risk early warning, the invention is characterized in that: the calculation process of the delfei-entropy method in the process three is as follows:
when calculating index entropy, firstly, respectively calculating the reject ratio of each specific detection parameter and the specific gravity of reject ratio of each specific detection parameter of the i-th detection item;
the calculation formula of the reject ratio proportion is as follows:
the calculation formula of the disqualification degree proportion is as follows:
the information entropy formula of the reject ratio in the i-th detection item is as follows:
the information entropy formula of the disqualification degree in the i-th detection item is as follows:
the information utility value of a certain index depends on the information entropy e of the index j The difference between the information utility value and 1 directly influences the weight, and the larger the information utility value is, the larger the importance of the information utility value to the evaluation is, and the larger the weight is;
the weight of the reject ratio in the i-th detection item is W 1 :
The weight of the disqualification degree in the i-th detection item is W 2 :
As a preferred scheme of the risk assessment model construction method for import food risk early warning, the invention is characterized in that: in the fifth flow, the first formula is as follows:
W Y =1-W X 。
as a preferred scheme of the risk assessment model construction method for import food risk early warning, the invention is characterized in that: the risk level and the proportion of the detection items in the process six are obtained by the risk value R of the first-level index, and the detection items are according to the R of the third-level index ij The values are determined by the big-to-order.
As a preferred scheme of the risk assessment model construction method for import food risk early warning, the invention is characterized in that: the specific process of constructing the fuzzy comprehensive evaluation model based on the AHP-risk matrix is as follows:
process one: designing a questionnaire, and inviting industry experts to score indexes from the two angles of possibility and severity of risk occurrence;
and a second process: according to expert investigation results, averaging the results of scoring the severity of the third-level indexes by the expert, and then carrying out assignment evaluation on the third-level indexes by adopting an improved Saath 1-9 scale method to obtain a judgment matrix of the third-level indexes, so that the situation that consistency test cannot pass can be avoided;
and a third process: carrying out weighted average on the severity score result of the third-level index and the weight thereof to obtain the severity score of the second-level index, and carrying out assignment evaluation on the index according to the improved Saath 1-9 scale method to obtain a judgment matrix of the second-level index;
and a process IV: after the weight vector is obtained by the AHP method, carrying out comprehensive operation with the established fuzzy membership matrix to obtain a fuzzy comprehensive evaluation vector;
and a fifth process: the fuzzy comprehensive evaluation vectors of the third-level indexes form a fuzzy membership matrix of the second-level indexes, and the fuzzy comprehensive evaluation vectors of the second-level indexes form a fuzzy membership matrix of the first-level indexes;
and a sixth process: multiplying the fuzzy comprehensive evaluation vector A of the first-level index by the evaluation level weighting value M to obtain a comprehensive score F, so as to judge the risk level of the food and the item to be detected;
process seven: meanwhile, the model result is compared with the evaluation result based on the laboratory detection data, and the parameters of the evaluation model based on the laboratory detection data can be modified and optimized.
Compared with the prior art:
the imported food risk assessment model established by the invention provides a new mode for sharing, fusing and excavating the imported food quality safety information, and can effectively identify related information such as food products, production enterprises, export countries/regions and the like with high risk, so that imported food safety risk control measures are more targeted, scientific and reasonable.
Drawings
FIG. 1 is a flow chart of a conventional analytic hierarchy process of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.
The invention provides a risk assessment model construction method for import food risk early warning, referring to fig. 1, comprising the following specific steps:
step one: based on big data technology and machine learning algorithm, through analyzing import food declaration data, qualified evaluation data, laboratory inspection data, risk warning notification information of related institutions/departments and social public opinion related safety risk information, a Hadoop/Spark architecture and an ETL tool are adopted to realize real-time or quasi-real-time acquisition of multi-source and multi-type data, and an HBase distributed database is adopted to store and retrieve mass multi-type data;
step two: the python is used as a basic tool for big data modeling, a scientific and effective imported food risk assessment model is built on the basis of a scikit-learn/tensorsurface machine learning/deep learning framework, related information such as imported food products with high risk, food production enterprises, export countries and the like is identified through the model, related management departments are assisted in carrying out scientific and effective imported food admission safety management, the imported food safety management level of China is integrally improved, 2 different models are adopted for assessing risk on imported food, a comprehensive assessment model based on laboratory detection data is constructed for laboratory data, and a fuzzy comprehensive assessment model based on an AHP-risk matrix is constructed for laboratory data-free.
The specific flow of the construction of the comprehensive evaluation model based on laboratory detection data is as follows:
the process is as follows: comparing the laboratory detection result with a standard prescribed limit value, respectively calculating to obtain the reject ratio and reject ratio of the third-level index, the second-level index and the first-level index, constructing a comprehensive evaluation model based on laboratory detection data, evaluating the two indexes by adopting the reject ratio and the reject ratio, and comparing the laboratory detection result with the standard prescribed limit value;
wherein, the reject ratio reflects the number of times of reject occurrence, and the reject ratio reflects the degree of deviation of the detection value from the standard limit value;
a second flow: after the reject ratio and reject ratio of the three-level indexes are calculated, a weighting method is adopted to obtain the risk value R of each level of indexes, wherein the larger the risk value R is, the higher the risk is;
and a process III: the weight is calculated by adopting a Deltaffy-entropy method, wherein the weight is calculated according to the discrete degree of the index by means of the concept of entropy in the Deltaffy-entropy method, and the reject ratio weight calculated by using the Deltaffy-entropy method are respectively
The calculation process of the delfei-entropy method is as follows:
when calculating index entropy, firstly, respectively calculating the reject ratio of each specific detection parameter and the specific gravity of reject ratio of each specific detection parameter of the i-th detection item;
the calculation formula of the reject ratio proportion is as follows:
the calculation formula of the disqualification degree proportion is as follows:
the information entropy formula of the reject ratio in the i-th detection item is as follows:
the information entropy formula of the disqualification degree in the i-th detection item is as follows:
the information utility value of a certain index depends on the information entropy e of the index j The difference between the information utility value and 1 directly influences the weight, and the larger the information utility value is, the larger the importance of the information utility value to the evaluation is, and the larger the weight is;
the weight of the reject ratio in the i-th detection item is W 1 :
The weight of the disqualification degree in the i-th detection item is W 2 :
The process is four: in order to avoid the weight distortion of the Delphi-entropy method, a Delphi expert scoring method is combined, and the reject ratio and the reject weight obtained by the Delphi method are respectivelyCalculating the weights of the two indexes of the reject ratio and the reject ratio by adopting a Delphi-entropy method, judging the discrete degree of a certain index by using an entropy value, and if the discrete degree of the index is larger, indicating that the index is more unstable, increasing the attention degree, wherein the weight calculated by using the entropy method is also larger;
the fifth flow is: calculating final weight W of reject ratio and reject ratio by using formula I X 、W Y ;
Wherein, formula one is as follows:
W Y =1-W X ;
the sixth flow: after the reject ratio, reject ratio and weight value of each level of index are obtained, the risk value of each level of index is obtained by using a weighted summation mode, so that the risk level and the detection item are determined;
wherein the risk level and the proportion of the detection items are obtained by the risk value R of the first-level index, and the detection items are according to the R of the third-level index ij The values are determined by the big-to-order.
In the traditional analytic hierarchy process, the judgment matrix is obtained by carrying out pairwise comparison on the index importance by an expert according to a Saath 1-9 scale method, when the evaluation index is too many, for example, 100 items of detection items related in a third-level index of the dairy product reach more than 100X 100, the used matrix is too large in scale, if judgment of 1/9 to 9 is required, the workload is too large, and the situation that consistency test cannot pass easily occurs.
The specific process of constructing the fuzzy comprehensive evaluation model based on the AHP-risk matrix is as follows:
process one: designing a questionnaire, and inviting industry experts to score indexes from the two angles of possibility and severity of risk occurrence;
and a second process: according to expert investigation results, averaging the results of scoring the severity of the third-level indexes by the expert, and then carrying out assignment evaluation on the third-level indexes by adopting an improved Saath 1-9 scale method to obtain a judgment matrix of the third-level indexes, so that the situation that consistency test cannot pass can be avoided;
and a third process: carrying out weighted average on the severity score result of the third-level index and the weight thereof to obtain the severity score of the second-level index, and carrying out assignment evaluation on the index according to the improved Saath 1-9 scale method to obtain a judgment matrix of the second-level index;
and a process IV: after the weight vector is obtained by the AHP method, carrying out comprehensive operation with the established fuzzy membership matrix to obtain a fuzzy comprehensive evaluation vector;
and a fifth process: the fuzzy comprehensive evaluation vectors of the third-level indexes form a fuzzy membership matrix of the second-level indexes, and the fuzzy comprehensive evaluation vectors of the second-level indexes form a fuzzy membership matrix of the first-level indexes;
and a sixth process: the fuzzy comprehensive evaluation vector A of the first-level index and the evaluation level weighting value M are multiplied to obtain a comprehensive score F, so that the risk level of the food and the item to be detected are judged, wherein the proportion of the risk level to the detection item is obtained by the F value of the first-level index, and the detection item is obtained according to the third-level indexThe values are determined from the order of magnitude (similar to the evaluation model based on laboratory test data, for example, when the F value of sterilized milk ranges from 60 to 80, the risk level is judged to be a medium risk, the ratio of test items is 70%, and the third level index (namely, specific test item) can be detected to be a medium risk>Items of 70% before the discharge, such as total colony count, risk values of salmonella, mould and the like, which need to be detected, and 30% after the discharge of lead, arsenic, aureomycin and the like, which need not to be detected);
process seven: meanwhile, the model result is compared with the evaluation result based on the laboratory detection data, and the parameters of the evaluation model based on the laboratory detection data can be modified and optimized.
Although the invention has been described hereinabove with reference to embodiments, various modifications thereof may be made and equivalents may be substituted for elements thereof without departing from the scope of the invention. In particular, the features of the disclosed embodiments may be combined with each other in any manner as long as there is no structural conflict, and the exhaustive description of these combinations is not given in this specification merely for the sake of omitting the descriptions and saving resources. Therefore, it is intended that the invention not be limited to the particular embodiment disclosed, but that the invention will include all embodiments falling within the scope of the appended claims.
Claims (7)
1. The risk assessment model construction method for import food risk early warning is characterized by comprising the following specific steps:
step one: based on big data technology and machine learning algorithm, through analyzing import food declaration data, qualified evaluation data, laboratory inspection data, risk warning notification information of related institutions/departments and social public opinion related safety risk information, a Hadoop/Spark architecture and an ETL tool are adopted to realize real-time or quasi-real-time acquisition of multi-source and multi-type data, and an HBase distributed database is adopted to store and retrieve mass multi-type data;
step two: the python is used as a basic tool for big data modeling, a scientific and effective imported food risk assessment model is built on the basis of a scikit-learn/tensorflow machine learning/deep learning framework, 2 different models are adopted for assessing risk on imported food, a comprehensive evaluation model with laboratory data is built on the basis of laboratory detection data, and a fuzzy comprehensive evaluation model without laboratory data is built on the basis of an AHP-risk matrix.
2. The method for constructing a risk assessment model for early warning of imported food risk according to claim 1, wherein the specific flow of the comprehensive assessment model construction based on laboratory detection data is as follows:
the process is as follows: comparing the laboratory detection result with a standard specified limit value, and respectively calculating to obtain the reject ratio and reject ratio of the third-level index, the second-level index and the first-level index;
a second flow: after the reject ratio and reject ratio of the three-level indexes are calculated, a weighting method is adopted to obtain the risk value R of each level of indexes, wherein the larger the risk value R is, the higher the risk is;
and a process III: by usingThe weight is calculated by a Deltaffy-entropy method, wherein the weight is calculated according to the discrete degree of the index by means of the concept of entropy in the Deltaffy-entropy method, and the reject ratio weight calculated by the Deltaffy-entropy method are respectively
The process is four: in order to avoid the weight distortion of the Delphi-entropy method, a Delphi expert scoring method is combined, and the reject ratio and the reject weight obtained by the Delphi method are respectively
The fifth flow is: calculating final weight W of reject ratio and reject ratio by using formula I X 、W Y ;
The sixth flow: after the reject ratio, reject ratio and weight value of each level of index are obtained, the risk value of each level of index is obtained by using a weighted summation mode, so that the risk level and the detection item are determined.
3. The method for constructing a risk assessment model for risk early warning of imported food according to claim 2, wherein the failure rate reflects the number of occurrences of failure, and the failure degree reflects the degree to which the detected value deviates from the standard limit value.
4. The method for constructing a risk assessment model for early warning of risk of imported food according to claim 2, wherein the calculation process of the delfei-entropy method in the third procedure is as follows:
when calculating index entropy, firstly, respectively calculating the reject ratio of each specific detection parameter and the specific gravity of reject ratio of each specific detection parameter of the i-th detection item;
the calculation formula of the reject ratio proportion is as follows:
the calculation formula of the disqualification degree proportion is as follows:
the information entropy formula of the reject ratio in the i-th detection item is as follows:
the information entropy formula of the disqualification degree in the i-th detection item is as follows:
the information utility value of a certain index depends on the information entropy e of the index j The difference between the information utility value and 1 directly influences the weight, and the larger the information utility value is, the larger the importance of the information utility value to the evaluation is, and the larger the weight is;
the weight of the reject ratio in the i-th detection item is W 1 :
The weight of the disqualification degree in the i-th detection item is W 2 :
5. The method for constructing a risk assessment model for risk early warning of imported food according to claim 2, wherein the first formula in the fifth flow is as follows:
W Y =1-W X 。
6. the method for constructing a risk assessment model for early warning of risk of imported food according to claim 2, wherein the risk level and the proportion of the detection items in the process six are obtained from the risk value R of the first level indicator, and the detection items are according to the R of the third level indicator ij The values are determined by the big-to-order.
7. The method for constructing a risk assessment model for early warning of imported food risk according to claim 1, wherein the specific process of constructing the fuzzy comprehensive assessment model based on the AHP-risk matrix is as follows:
process one: designing a questionnaire, and inviting industry experts to score indexes from the two angles of possibility and severity of risk occurrence;
and a second process: according to expert investigation results, averaging the results of scoring the severity of the third-level indexes by the expert, and then carrying out assignment evaluation on the third-level indexes by adopting an improved Saath 1-9 scale method to obtain a judgment matrix of the third-level indexes, so that the situation that consistency test cannot pass can be avoided;
and a third process: carrying out weighted average on the severity score result of the third-level index and the weight thereof to obtain the severity score of the second-level index, and carrying out assignment evaluation on the index according to the improved Saath 1-9 scale method to obtain a judgment matrix of the second-level index;
and a process IV: after the weight vector is obtained by the AHP method, carrying out comprehensive operation with the established fuzzy membership matrix to obtain a fuzzy comprehensive evaluation vector;
and a fifth process: the fuzzy comprehensive evaluation vectors of the third-level indexes form a fuzzy membership matrix of the second-level indexes, and the fuzzy comprehensive evaluation vectors of the second-level indexes form a fuzzy membership matrix of the first-level indexes;
and a sixth process: multiplying the fuzzy comprehensive evaluation vector A of the first-level index by the evaluation level weighting value M to obtain a comprehensive score F, so as to judge the risk level of the food and the item to be detected;
process seven: meanwhile, the model result is compared with the evaluation result based on the laboratory detection data, and the parameters of the evaluation model based on the laboratory detection data can be modified and optimized.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310669481.4A CN116703158A (en) | 2023-06-06 | 2023-06-06 | Risk assessment model construction method for imported food risk early warning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310669481.4A CN116703158A (en) | 2023-06-06 | 2023-06-06 | Risk assessment model construction method for imported food risk early warning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116703158A true CN116703158A (en) | 2023-09-05 |
Family
ID=87838685
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310669481.4A Pending CN116703158A (en) | 2023-06-06 | 2023-06-06 | Risk assessment model construction method for imported food risk early warning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116703158A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117035563A (en) * | 2023-10-10 | 2023-11-10 | 河北省产品质量监督检验研究院 | Product quality safety risk monitoring method, device, monitoring system and medium |
-
2023
- 2023-06-06 CN CN202310669481.4A patent/CN116703158A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117035563A (en) * | 2023-10-10 | 2023-11-10 | 河北省产品质量监督检验研究院 | Product quality safety risk monitoring method, device, monitoring system and medium |
CN117035563B (en) * | 2023-10-10 | 2023-12-26 | 河北省产品质量监督检验研究院 | Product quality safety risk monitoring method, device, monitoring system and medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111461555B (en) | Production line quality monitoring method, device and system | |
CN105117602B (en) | A kind of metering device running status method for early warning | |
CN112199367B (en) | Long-time scale power balance check and big data analysis method and system | |
CN106651169A (en) | Fuzzy comprehensive evaluation-based distribution automation terminal state evaluation method and system | |
CN103337043B (en) | The method for early warning of electric power communication device running status and system | |
CN108898311A (en) | A kind of data quality checking method towards intelligent distribution network repairing dispatching platform | |
CN111008193B (en) | Data cleaning and quality evaluation method and system | |
CN110222991B (en) | Metering device fault diagnosis method based on RF-GBDT | |
CN116703158A (en) | Risk assessment model construction method for imported food risk early warning | |
CN112381364A (en) | Comprehensive evaluation method for food quality spot check | |
CN109146611A (en) | A kind of electric business product quality credit index analysis method and system | |
CN115358155A (en) | Power big data abnormity early warning method, device, equipment and readable storage medium | |
CN111598165A (en) | Density clustering outlier detection method based on extreme learning machine | |
Zhang | Food safety risk intelligence early warning based on support vector machine | |
Li et al. | Dynamic failure mode analysis approach based on an improved Taguchi process capability index | |
CN115619289A (en) | Block chain-based distribution transformer quality detection result distinguishing method and device | |
Wang et al. | E‐Commerce Credit Risk Assessment Based on Fuzzy Neural Network | |
CN111080466A (en) | Calender calendering quality early warning system based on big data | |
Pati et al. | Statistical quality study of the parts produced in an automobile industry: a daimler India case study | |
CN113762764A (en) | Automatic grading and early warning system and method for safety risk of imported food | |
CN101706898A (en) | Europe-exported consumer goods security risk assessment early warning system and method for inspection and quarantine | |
Saeed et al. | Electricity theft detection in power utilities using bagged CHAID-based classification trees | |
CN115392618A (en) | AHP-EW and AE-RNN fusion-based food safety risk early warning model and establishment method thereof | |
CN115187026A (en) | Industrial risk monitoring method and system and readable storage medium | |
Xu et al. | Static Correlation Analysis with Industrial Electricity Consumption Volumes |
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 |