Dairy product food safety risk factor data collection and quantification method

Abstract

The invention discloses a dairy product food safety risk factor data collection and quantification method, and belongs to the field of dairy product food safety, and the method comprises the steps: defining risk factors of collected data, deploying related Internet of Things equipment, collecting and transmitting data in real time, preprocessing and cleaning the data, quantifying the risk factor data, and analyzing and feeding back the data. Quantitative or qualitative quantitative processing is carried out on the risk factor data to obtain quantitative data which can be used for data analysis, and potential food safety problems are further found and fed back in real time, so that the production process is adjusted and optimized, and food safety is guaranteed. The method has the characteristics of being real-time, accurate and comprehensive, food safety risk control of various links of production, processing, packaging, storage and transportation of the dairy products is facilitated, and the efficiency and effect of food safety management of the dairy products are greatly improved.

Description

Technical Field

[0001] The present invention belongs to the field of dairy product safety, and more specifically relates to a method for collecting and quantifying dairy product food safety risk factor data. Background Art

[0002] Dairy products are one of the important foods in people's daily lives, and their food safety issues have always been the focus of people's attention. Especially in recent years, the core requirement for food safety is prevention-oriented, emphasizing early warning and preventive control of food safety risks. The production process of dairy products involves multiple links such as production, processing, packaging, storage, and transportation. There are many food safety risk factors, including microbial contamination, irregular operation, equipment operation status, raw material quality, supply chain management, product inspection and quarantine, etc. Therefore, how to effectively collect and quantify these risk factors so as to accurately predict food safety risks is the key to food safety risk management.

[0003] Traditional food safety risk factor analysis methods mainly include biological risk assessment, chemical risk assessment and physical risk assessment, etc. These assessment methods are mainly based on laboratory experimental data, and their experimental process is complex, long, and costly, and cannot meet the needs of early warning and preventive control of food safety risks. Therefore, there is an urgent need for a method that can collect and quantify food safety risk factor data in real time, accurately and effectively.

[0004] With the development and application of Internet of Things technology, it has realized the intelligent collection and processing of data in many fields, including food safety. However, the application of existing Internet of Things technology in the collection and quantification of food safety risk factor data is still insufficient, including unclear definition of risk factor data, unscientific data collection and preprocessing methods, weak data analysis and feedback capabilities, etc. Therefore, a new method for collecting and quantifying dairy food safety risk factor data is urgently needed to solve these problems. Summary of the invention

[0005] The present invention aims to solve the problems in the prior art such as unclear definition of dairy food safety risk factor data, data collection and processing methods not meeting real-time requirements, inaccurate quantification of risk factors, and weak data analysis and feedback capabilities. A new method for collecting and quantifying dairy food safety risk factor data is proposed, which can collect risk factor data in real time, accurately and effectively perform preprocessing and quantification, scientifically analyze risks and provide timely feedback, thereby improving the food safety risk early warning and prevention capabilities of dairy products.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: the method comprises:

[0007] Define the risk factors for data collection: including microbial contamination and improper operation; the data collected mainly include storage temperature, humidity, and pH value;

[0008] Deploy relevant IoT devices: deploy corresponding sensors and devices in the production, processing, packaging, storage and transportation of dairy products;

[0009] Real-time data collection and transmission: Use the Internet of Things to receive data collected from various sensors and devices in real time and quickly transmit it to the database or cloud server;

[0010] Data preprocessing and cleaning: preliminarily process the collected raw data, clean up redundant, invalid or erroneous data, process missing data, and convert data formats;

[0011] Quantification of risk factor data: Quantify the cleaned data.

[0012] In one embodiment, the deployment of related IoT devices includes

[0013] Select appropriate sensors and equipment based on the identified risk factor data collection requirements;

[0014] Install sensors and equipment according to equipment user manuals;

[0015] Connect sensors and devices to the IoT platform and exchange data with the platform.

[0016] In one embodiment, the quantification of risk factor data includes quantification of qualitative factors and quantification of quantitative factors.

[0017] In one embodiment, the quantification of the quantitative factor comprises:

[0018] Risk factors include "extremely large" factors, "extremely small" factors, "medium" factors and "interval" factors. In the risk analysis model, the larger the value of the "extremely large" factor, the higher the risk.

[0019] Positive factors, microbial contaminants, i.e. factors with greater risk as the factor value increases, the dimensionless processing method is as follows;

[0020]

[0021] In the formula: y is the evaluation value of the factor, x is the actual value of the dimensionless factor, max is the maximum value of the dimensionless factor, x min is the minimum value of the dimensionless factor;

[0022] The inverse factor, pH value, is the factor with greater risk the smaller the value of this factor. The dimensionless processing method is as follows;

[0023]

[0024] Interval factors include storage temperature and humidity. Interval factors refer to factors that change within a certain range. The risk level has no direct relationship with the level. Instead, the further away from the optimal range, the higher the risk level. The treatment method is as follows:

[0025]

[0026] In one embodiment, the quantification of the qualitative factors includes:

[0027] Qualitative factors include the standardization of production operations: equipment operation status: raw material quality: supply chain management: product inspection and quarantine;

[0028] Fuzzy theory is used for quantification.

[0029] In one solution, the specific process of quantifying the qualitative factors is as follows: constructing a judgment set of basic elements of the qualitative factors and risk degree and risk level, wherein the judgment set includes five levels, namely, "low, lower, medium, higher, and high", and obtaining the corresponding relationship in the risk value table according to the corresponding relationship in the table, and mapping each row;

[0030] The mapping steps are as follows:

[0031] (1) The first column is for the “low” level, the second column is for the “lower” level, the third column is for the “medium” level, the fourth column is for the “higher” level, and the fifth column is for the “high” level;

[0032] (i.e. w1 = "low", w2 = "lower", w3 = "medium", w4 = "higher", w5 = "high")

[0033] (2) Count the number of risk factors belonging to each level in each row, denoted as wj (j = 1, 2, ..., 5), where represents the number of risk factors belonging to level j in the row;

[0034] (3) Construct the risk membership matrix;

[0035] Let K represent the risk matrix:

[0036]

[0037] Assign values ​​to the risk membership matrix:

[0038]

[0039] Where n represents the number of experts. At this point, the secondary fuzzification is completed and the risk membership matrix can be obtained.

[0040] The entropy coefficient is calculated according to the value in the risk membership matrix. The entropy coefficient is calculated using the entropy coefficient method. The entropy coefficient of each risk factor is calculated to obtain the entropy coefficient Qi, that is, the weight of each risk factor M = (m 1 , m 2 ,…,m n ), Qi is one of them i , assign corresponding weights to each i element in the evaluation set, and obtain the weight vector N of the evaluation set = (n 1 , n 2 ,…,n n ), where n is the number of elements in the risk assessment set; the calculated risk value is:

[0041] R=M·K·N T .

[0042] In one scenario, the corresponding relationship between risk value and risk level is as follows:

[0043] R is 0-0.2, risk level: low;

[0044] R is 0.2-0.4, risk level: low;

[0045] R is 0.4-0.6, risk level: medium;

[0046] R is 0.6-0.8, risk level: high;

[0047] R is 0.8-1, and the risk level is high.

[0048] Beneficial effects of the present invention:

[0049] The beneficial effects of the present invention are mainly reflected in the realization of real-time collection and transmission of dairy food safety risk factor data, improving the efficiency and timeliness of data acquisition, and proposing a preprocessing and cleaning method for risk factor data, providing an accurate basis for subsequent risk analysis. In addition, the method scientifically quantifies the risk factor data and more accurately describes each risk factor, thereby enhancing the accuracy of risk analysis. Most importantly, the present invention realizes real-time analysis and feedback of risk factor data, which can promptly discover and deal with potential food safety issues, and effectively support food safety management and decision-making. Overall, the method can effectively improve the safety of dairy foods, reduce food safety risks, and thus protect the health of consumers. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 is a flow chart of the method of the present invention;

[0051] Figure 2 This is the specific process of quantifying the qualitative factors of the present invention. DETAILED DESCRIPTION

[0052] In order to facilitate understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. Typical embodiments of the present invention are provided in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described in the present invention. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0053] Unless otherwise defined, all technical and scientific terms used in the present invention have the same meanings as those understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0054] like Figure 1 As shown, the method for collecting and quantifying data on food safety risk factors of dairy products comprises:

[0055] S1. Define the risk factors for collecting data: including microbial contamination and improper operation; the collected data mainly include storage temperature, humidity, and pH value.

[0056] According to international food safety standards and the characteristics of dairy products, the types of risk factors that may affect the safety of dairy products are determined, mainly including microbial contamination and irregular operations in the production process; for each risk factor, indicators and parameters that can reflect its risk level are defined. Microbial contamination can be measured by indicators such as microbial species, concentration, and growth rate. The frequency of data collection is determined based on the required accuracy and actual conditions. For environmental factors such as temperature and humidity, data needs to be collected every few minutes; for operating specifications, data is collected once at the beginning and end of each operation.

[0057] S2. Deploy relevant IoT devices: Deploy corresponding sensors and equipment in the production, processing, packaging, storage and transportation of dairy products.

[0058] Select the appropriate sensors and equipment: This step needs to be carried out according to the risk factor data collection requirements determined in S1. To collect temperature and humidity data, you need to select a temperature and humidity sensor; to collect pH value data, select a pH value sensor; to track the logistics process of dairy products, select a location tracker and other equipment.

[0059] Install the equipment according to the equipment user manual. Temperature and humidity sensors should be placed in key links where temperature and humidity changes may affect the quality of dairy products, such as cooling, fermentation, maturation, storage, etc. At the same time, the equipment needs to be configured according to the equipment function and data collection requirements. Set the data collection frequency, alarm threshold, etc. of the sensor.

[0060] Connect the device to the IoT platform to enable data exchange between the device and the platform, including selecting the appropriate communication protocol (such as NB-IoT, LoRa, ZigBee, etc.), entering the device ID and related authorization information to register the device, and setting the data upload path.

[0061] After the equipment is installed and connected, a running test should be performed to verify whether the equipment functions normally and whether data collection and uploading meet the requirements. If any problems are found, adjustments need to be made in a timely manner.

[0062] S3. Real-time data collection and transmission: Use the Internet of Things to receive data collected from various sensors and devices in real time, and quickly transmit it to the database or cloud server.

[0063] According to the various IoT devices deployed in S2, the data monitored by each sensor is collected in real time according to the set frequency. The IoT devices transmit the collected data to the database or cloud server in real time through cloud services. Data transmission requires reliable network links, such as Wi-Fi, 4G or 5G.

[0064] S4. Data preprocessing and cleaning: Perform preliminary processing on the collected raw data, clean up redundant, invalid or erroneous data, process missing data, and convert data formats.

[0065] Check whether the data contains redundant, abnormal, erroneous or missing values. If so, they need to be processed. For example, redundant data can be deleted, abnormal or erroneous data can be corrected, and missing data can be filled.

[0066] Since the data comes from different sensors and devices, there are data deviations or inconsistencies. In order to ensure the consistency of the data, the data needs to be standardized, such as unit conversion, data format conversion, etc.

[0067] Integrate the data collected from different sensors and devices. For multiple risk factor data collected at the same time point, group them into the same record to facilitate subsequent data analysis and risk assessment.

[0068] S5. Quantification of risk factor data: Quantify the cleaned data.

[0069] The risk factor data quantification includes the quantification of qualitative factors and the quantification of quantitative factors.

[0070] The quantification of the quantitative factors includes: the quantitative risk factors contain "extremely large" factors, "extremely small" factors, "medium" factors and "interval" factors. The "extremely large" factor means that the larger the value of the factor, the higher the risk, and so on for other factors. At the same time, due to the differences in the units and orders of magnitude between the quantitative factors, there is incommensurability, which brings inconvenience to the comparison of the size of the comprehensive evaluation factors. Therefore, it is necessary to process the factors dimensionlessly, use certain measurement methods to eliminate the differences in measurement units and values ​​between the factors, convert the measured values ​​into factor evaluation values, that is, calculate the risk factor evaluation value. In addition to knowing the measured values ​​of the risk factors, it is also necessary to determine the maximum and minimum values.

[0071] Positive factors, microbial contaminants, i.e. factors with greater risk as the factor value increases, the dimensionless processing method is as follows;

[0072]

[0073] In the formula: y is the evaluation value of the factor, x is the actual value of the dimensionless factor, max is the maximum value of the dimensionless factor, x min is the minimum value of the dimensionless factor;

[0074] The inverse factor, pH value, is the factor with greater risk the smaller the value of this factor. The dimensionless processing method is as follows;

[0075]

[0076] Interval factors include storage temperature and humidity. Interval factors refer to factors that change within a certain range. The risk level has no direct relationship with the level. Instead, the further away from the optimal range, the higher the risk level. The treatment method is as follows:

[0077]

[0078] According to the above method, the quantitative factors can be dimensionlessly processed respectively.

[0079] The quantification of the qualitative factors includes:

[0080] Qualitative factors include the standardization of production process operations, equipment operation status, raw material quality, supply chain management, and product inspection and quarantine. Improved fuzzy mathematics methods are used to quantify fuzzy information.

[0081] like Figure 2 As shown, the specific process of quantifying the qualitative factors is as follows:

[0082] S501. Construct the basic elements of qualitative factors and the evaluation set of risk degree and risk level. The evaluation set includes five levels, namely "low, lower, medium, higher, and high". According to the corresponding relationship in the table, the corresponding corresponding relationship in the risk value table is obtained, and each row is mapped;

[0083] S502, the mapping steps are as follows:

[0084] (1) The first column is for the “low” level, the second column is for the “lower” level, the third column is for the “medium” level, the fourth column is for the “higher” level, and the fifth column is for the “high” level;

[0085] (i.e. w1 = "low", w2 = "lower", w3 = "medium", w4 = "higher", w5 = "high")

[0086] (2) Count the number of risk factors belonging to each level in each row, denoted as wj (j = 1, 2, ..., 5), where represents the number of risk factors belonging to level j in the row;

[0087] (3) Construct the risk membership matrix;

[0088] Let K represent the risk matrix:

[0089]

[0090] Assign values ​​to the risk membership matrix:

[0091]

[0092] Where n represents the number of experts. At this point, the secondary fuzzification is completed and the risk membership matrix can be obtained.

[0093] S503, calculate the entropy coefficient according to the value in the risk membership matrix. Calculate the entropy coefficient of each risk factor to obtain the entropy coefficient Qi, that is, the weight of each risk factor M = (m 1 , m 2 ,…,m n ), Qi is one of them i , assign corresponding weights to each i element in the evaluation set, and obtain the weight vector N of the evaluation set = (n 1 , n 2 ,…,n n ), where n is the number of elements in the risk assessment set; the calculated risk value is:

[0094] R=M·K·N T .

[0095] The corresponding relationship between risk value and risk level is as follows:

[0096] R is 0-0.2, risk level: low;

[0097] R is 0.2-0.4, risk level: low;

[0098] R is 0.4-0.6, risk level: medium;

[0099] R is 0.6-0.8, risk level: high;

[0100] R is 0.8-1, and the risk level is high.

[0101] Embodiment 1:

[0102] Assume that a dairy product manufacturing company uses the dairy food safety risk factor data collection and quantification method. In order to ensure food safety and minimize food safety risks, they define the risk factors for collecting data, including microbial contamination (such as Salmonella, Listeria, etc.) and irregular operations. The collected data also includes storage temperature (0-4°C), humidity (75%-85%), and pH value (6.5-6.7).

[0103] On the production line, various IoT devices are deployed, including temperature sensors, humidity sensors and pH value detection equipment. These sensors will continuously collect data and transmit it to the company's database or cloud server in real time via the Internet.

[0104] When data enters the database, data preprocessing and cleaning are performed, including cleaning up redundant data, invalid data or erroneous data, processing missing data, converting data formats, and other steps to ensure the accuracy and completeness of the data.

[0105] Next is the quantification of risk factor data, which is rare in the industry. For example, microbial contamination is a positive factor, and its presence will lead to increased risk.

[0106] The next step is to quantify qualitative factors, such as operational norms, equipment operating conditions, raw material quality, supply chain management, and product inspection and quarantine. They use fuzzy theory for quantification. The specific process includes constructing the basic elements of qualitative factors and the judgment set of risk degree and risk level, and then quantifying them into numbers. The detailed quantification process is omitted.

[0107] Finally, they will analyze the quantified risk factor data, study the relationship between risk factors, identify potential food safety issues, and provide real-time feedback to relevant employees to guide them in optimizing operations and preventing possible risks.

[0108] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM).

[0109] It should be understood that the detailed description of the technical solutions of the present invention by means of the preferred embodiments is illustrative rather than restrictive. A person skilled in the art may modify the technical solutions described in the embodiments, or replace some of the technical features by equivalents, based on reading the specification of the present invention; and these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for collecting and quantifying data on dairy food safety risk factors, characterized in that: The method includes: Define the risk factors for data collection: including microbial contamination and improper operation; the data collected mainly include storage temperature, humidity, and pH value; Deploy relevant IoT devices: deploy corresponding sensors and devices in the production, processing, packaging, storage and transportation of dairy products; Real-time data collection and transmission: Use the Internet of Things to receive data collected from various sensors and devices in real time and quickly transmit it to the database or cloud server; Data preprocessing and cleaning: preliminarily process the collected raw data, clean up redundant, invalid or erroneous data, process missing data, and convert data formats; Quantification of risk factor data: Quantify the cleaned data.

2. The method for collecting and quantifying dairy food safety risk factor data according to claim 1, characterized in that: The deployment-related IoT devices include Select appropriate sensors and equipment based on the identified risk factor data collection requirements; Install sensors and equipment according to equipment user manuals; Connect sensors and devices to the IoT platform and exchange data with the platform.

3. The method for collecting and quantifying dairy food safety risk factor data according to claim 1, characterized in that: The risk factor data quantification includes the quantification of qualitative factors and the quantification of quantitative factors.

4. The method for collecting and quantifying dairy food safety risk factor data according to claim 3, characterized in that: The quantification of the quantitative factors includes: Risk factors include "extremely large" factors, "extremely small" factors, "medium" factors and "interval" factors. In the risk analysis model, the "extremely large" factor means that the larger the value of the factor, the higher the risk. Positive factors, microbial contaminants, i.e. factors with greater risk as the factor value increases, the dimensionless processing method is as follows; In the formula: y is the evaluation value of the factor, x is the actual value of the dimensionless factor, max is the maximum value of the dimensionless factor, x min is the minimum value of the dimensionless factor; The inverse factor, pH value, is the factor with greater risk the smaller the value of this factor. The dimensionless processing method is as follows; Interval factors include storage temperature and humidity. Interval factors refer to factors that change within a certain range. The risk level has no direct relationship with the level. Instead, the further away from the optimal range, the higher the risk level. The treatment method is as follows:

5. The method for collecting and quantifying dairy food safety risk factor data according to claim 3, characterized in that: The quantification of the qualitative factors includes: Qualitative factors include the standardization of production operations: equipment operation status: raw material quality: supply chain management: product inspection and quarantine; Improved fuzzy theory is used for quantification.

6. The method for collecting and quantifying data on dairy food safety risk factors according to claim 5, characterized in that: The specific process of quantifying the qualitative factors is as follows: constructing the basic elements of the qualitative factors and the evaluation set of risk degree and risk level, the evaluation set includes five levels, namely "low, lower, medium, higher, high", and according to the corresponding relationship in the table, obtaining the corresponding corresponding relationship in the risk value table, and mapping each row; The mapping steps are as follows: (1) The first column is for the "low" level, the second column is for the "lower" level, the third column is for the "medium" level, the fourth column is for the "higher" level, and the fifth column is for the "high" level; (i.e. w1 = "low", w2 = "lower", w3 = "medium", w4 = "higher", w5 = "high") (2) Count the number of risk factors belonging to each level in each row, denoted as wj (j = 1, 2, ..., 5), where represents the number of risk factors belonging to level j in the row; (3) Construct the risk membership matrix; Let K represent the risk matrix: Assign values ​​to the risk membership matrix: Where n represents the number of experts; At this point, the secondary fuzzification is completed and the risk membership matrix can be obtained; The entropy coefficient is calculated according to the value in the risk membership matrix. The entropy coefficient is calculated using the entropy coefficient method. The entropy coefficient of each risk factor is calculated to obtain the entropy coefficient Qi, that is, the weight of each risk factor M = (m1, m2, ..., m n ), Qi is one of them i , assign corresponding weights to each i element in the evaluation set, and obtain the weight vector N of the evaluation set = (n1, n2, ..., n n ), where n is the number of elements in the risk assessment set; the calculated risk value is: R=M·K·N T 。 7. A method for collecting and quantifying dairy food safety risk factor data according to claim 6, characterized in that: The corresponding relationship between risk value and risk level is as follows: R is 0-0.2, risk level: low; R is 0.2-0.4, risk level: low; R is 0.4-0.6, risk level: medium; R is 0.6-0.8, risk level: high; R is 0.8-1, and the risk level is high.

Images