Food quality safety tracking management system based on big data

Abstract

The invention discloses a food quality safety tracking management system based on big data, and relates to the technical field of food safety, the system can collect various quality data in real time at key nodes of dairy product production and storage, and realize automatic association of batch information. The parameter acquisition module acquires dairy product production data in real time through acquisition equipment installed at a production node, and acquires storage data at a storage node at the same time. After the data is transmitted to the central database, the system preprocesses the data, and finally a complete preservative traceability data set is generated. And associating the quality data of each batch by using the unique identifier, and associating the dairy product batch ID with the preservative batch ID to form a dairy product component association table PCA, and establishing a detailed tracing relationship between different batches. The module enhances the traceability and transparency of the product quality, and is convenient for quickly finding out the problem source when the quality problem occurs.

Description

Technical Field

[0001] The present invention relates to the technical field of food safety, and in particular to a food quality and safety tracking management system based on big data. Background Art

[0002] As food safety supervision becomes increasingly intelligent, the field of dairy product quality management relies more and more on information systems. Tracking and analyzing the life cycle of additives, especially preservatives, used in the production, processing, storage and circulation of dairy products through big data technology has become one of the key links in quality traceability. The life cycle data of ingredients such as preservatives directly affects the shelf life, quality stability and safety of dairy products. Therefore, how to accurately collect, analyze and track the stability and effect attenuation of these ingredients under different environments and storage conditions has become an important innovation direction for dairy product quality management.

[0003] At the current stage, most existing systems only conduct fixed-frequency sampling inspections on finished dairy products, and are unable to dynamically monitor the actual effectiveness of preservatives and their changing trends under different conditions in real time. In addition, most systems lack a comprehensive analysis of the use cycle and environmental impact of preservatives, and are unable to provide data support such as ingredient stability decay and effect prediction. Due to the lack of ingredient life cycle management, many companies find it difficult to identify the reasons in a timely manner when faced with product quality fluctuations or shelf life that does not meet standards, thus affecting the efficiency of quality control. Current technology is still insufficient in how to establish the relationship between preservative use, environmental factors and the shelf life of dairy products, which makes it difficult to effectively predict and control the actual effects of preservatives and the quality of dairy products. Summary of the invention

[0004] In view of the deficiencies of the prior art, the present invention provides a food quality and safety tracking and management system based on big data, which solves the problems mentioned in the background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: including a parameter acquisition module, a link data association module, a preservative stability analysis module, a dairy product shelf life prediction module and a data visualization and report generation module;

[0006] The parameter acquisition module collects the quality data of the source node of dairy products through the data acquisition equipment, and transmits the preservative traceability data to the central database, and performs preprocessing to obtain the preservative traceability data set;

[0007] The link data association module manages the preservative traceability data set in a unified manner through the big data platform, and uses unique identification of multi-table data to form a complete dairy ingredient association table PCA;

[0008] The preservative stability analysis module establishes a component stability algorithm model in the central database, calculates and outputs the preservative stability index S(t) based on the preservative traceability data set, and further constructs a preservative effect attenuation integral model based on the output preservative stability index S(t), and calculates and outputs the preservative comprehensive effect index E(t);

[0009] The dairy product shelf life prediction module outputs a shelf life estimate L by analyzing the attenuation effect of the preservative comprehensive effect index E(t), and then evaluates the shelf life by comparing the shelf life estimate L with the set normal shelf life Tnorm. Based on the evaluation results, a multi-level early warning mechanism is generated to indicate the stability of the preservative.

[0010] The data visualization and report generation module generates visualization reports by building a visualization platform, and generates corresponding graphical presentations and analyses based on the visualization reports through visualization tools.

[0011] Preferably, the parameter acquisition module includes an acquisition unit and a preprocessing unit;

[0012] The collection unit collects the quality data of preservatives used in the production process of dairy products in real time by installing data collection equipment at the source node of dairy products;

[0013] The source node includes a production node and a storage node, and the quality data includes production data and storage data;

[0014] The production node is equipped with collection equipment including Ph sensors, microbial online monitoring equipment, near infrared spectrometer NIR, thermocouples and rotary encoders; the production data includes Ph change value △PH of dairy products, microbial content Ws, fat content Zf, protein content Db, processing temperature Tj and stirring rate R;

[0015] The storage node includes a temperature sensor, a humidity sensor and an oxygen concentration sensor through the installed collection equipment; the stored data includes the storage environment temperature T, the storage environment humidity H and the oxygen concentration O.

[0016] Preferably, the preprocessing unit is used to transmit the collected quality data to the central database by connecting the sensor to the central database through a network using an Internet of Things device, and preprocess the collected quality data in the central database, wherein the preprocessing method includes data cleaning and standardization, and after preprocessing, the quality data is timestamped and integrated according to the sensor acquisition time to obtain a preservative traceability data set, and is stored in the central database in the order of timestamps;

[0017] The preservative traceability data set includes a production data set and a storage data set;

[0018] The production data set includes the Ph change value △PH(t), microbial content Ws(t), fat content Zf(t) and protein content Db(t) of dairy products at time t respectively;

[0019] The storage data set includes the storage environment temperature T(t), the storage environment humidity H(t) and the oxygen concentration O(t) at time t.

[0020] Preferably, the link data association module includes a production management integration unit and a database association unit;

[0021] The production management integration unit tracks all preservatives used in dairy products through the big data platform, generates a preservative data table, records the supplier, generation date and preservative batch data, and then combines the data set stored in the central database with the preservative data table to generate a unique identifier, and integrates to obtain the preservative batch ID;

[0022] The preservative batch ID is composed of a 20-digit ID and a barcode, wherein four digits represent the supplier number, six digits represent the production date, four digits represent the batch number, two digits represent the storage environment temperature, two digits represent the storage environment humidity, and two digits represent the oxygen concentration;

[0023] During the process of adding preservatives, the preservative batch ID is entered into the production workshop management system, and automatically synchronized with the dairy product batch ID being produced into the central database, and the PCA of the associated dairy product ingredients is generated;

[0024] The dairy product batch ID is composed of a 16-digit ID, which includes a three-digit factory number, a four-digit production date, a three-digit production line number and a three-digit product type.

[0025] Preferably, the database association unit constructs a query rule through the dairy product component association table PCA, corresponds each dairy product batch ID to a preservative batch ID, and then sets a traceability mark according to the production date of the dairy product and the production date of the preservative;

[0026] The dairy product component association table PCA includes dairy product batch ID, preservative batch ID and preservative production date.

[0027] Preferably, the preservative stability analysis module includes a preservative stability analysis unit and a preservative effect attenuation unit;

[0028] The preservative stability analysis unit establishes a component stability algorithm model, extracts a stored data set and inputs it into the component stability algorithm model to output a preservative stability index S(t) of the preservative at time t;

[0029] The preservative stability index S(t) is calculated and output by the following component stability algorithm model;

[0030]

[0031] In the formula, e represents the exponential function, α represents the basic coefficient of the influence of temperature on stability, β represents the nonlinear degree of temperature influence, which is used to adjust the nonlinear effect of storage environment temperature, γ represents the attenuation coefficient of humidity on stability, and t represents the unit time t.

[0032] Preferably, the preservative effect attenuation unit constructs a preservative effect attenuation integral model through the preservative stability index S(t) output by the component stability algorithm model, calculates and outputs the preservative comprehensive effect index E(t) at time t, and analyzes the attenuation effect of the preservative during the shelf life of the dairy product;

[0033] The preservative comprehensive effect index E(t) is calculated and output by the following preservative effect attenuation integral model;

[0034]

[0035] Where κ represents the attenuation coefficient and dt represents the time-integrated variable.

[0036] Preferably, the dairy product shelf life prediction module includes a prediction analysis unit and a quality fluctuation early warning unit;

[0037] The prediction and analysis unit calculates and outputs a shelf life estimate L based on the obtained preservative comprehensive effect index E(t) in combination with the production data set, and analyzes the actual shelf life of the dairy product;

[0038] The estimated shelf life value L is calculated and outputted by the following algorithm formula;

[0039]

[0040] In the formula, Texp represents the expected shelf life, φ represents the influence coefficient of preservative effect on shelf life, ψ represents the influence coefficient of dairy raw material quality on shelf life, represents the quality of dairy raw materials, ω represents the influence coefficient of production process on shelf life, and (T(t)·R(t)) represents the dynamic production process parameters.

[0041] Preferably, the quality fluctuation warning unit evaluates the shelf life and analyzes the deviation between the predicted shelf life value and the expected standard value based on the obtained estimated shelf life value L and the normal shelf life Tnorm, and generates a multi-level warning mechanism based on the evaluation result. The specific evaluation content is as follows;

[0042] When the estimated shelf life value L ≥ the normal shelf life Tnorm, it means that the current batch of dairy products is within the standard shelf life and the quality is stable, and no warning needs to be generated;

[0043] When the estimated shelf life value L is less than 0.9×normal shelf life Tnorm, a first-level warning message is generated, suggesting that the preservative batch and raw material quality be checked;

[0044] When the estimated shelf life value L is less than 0.8×normal shelf life Tnorm, a second-level warning message is generated, suggesting further confirmation of process parameters and storage environment;

[0045] When the estimated shelf life value L is less than 0.7×normal shelf life Tnorm, a third-level warning message is generated, which automatically triggers the quality control process, suspends the current dairy product shipment, and prompts relevant quality control personnel to investigate the problem.

[0046] Preferably, the visualization and report generation module generates a visualization report by constructing a visualization platform, combining the collected quality data set and the output preservative stability index S(t) and shelf life estimate L, wherein the visualization report includes a preservative stability report SIR, a shelf life prediction report LPR and a process monitoring report PMR, and is graphically displayed through a visualization tool;

[0047] The preservative stability report SIR generates a line chart using data analysis tools and a visualization library to analyze the preservative stability index S(t) of multiple preservative batches, showing the preservative stability index S(t) curves of different preservative batches, and adds color annotations to the chart to mark the mutation points of the preservative stability index S(t);

[0048] The shelf life prediction report LPR generates the first-level warning information, second-level warning information and third-level information through the shelf life evaluation, and uses a time series graph to display the dairy product batches with warning information and the warning information prompt content;

[0049] The process monitoring report PMR uses a visualization tool to sort the production status of dairy products in time series based on the production data set, and displays it by using a line graph. At the same time, standard processing parameters are set to highlight abnormal dairy processing batches in the production data set.

[0050] The present invention provides a food quality and safety tracking management system based on big data. It has the following beneficial effects:

[0051] (1) Through the parameter collection module and the link data association module, the system can collect various quality data in real time at the key nodes of dairy production and storage, and realize the automatic association of batch information. The parameter collection module collects dairy production data in real time through pH sensors, microbial online monitoring equipment, near-infrared spectrometers (NIR) and other devices installed at the production nodes, and collects temperature, humidity and oxygen concentration data of the storage environment at the storage nodes. After the data is transmitted to the central database, the system will clean, standardize and timestamp the data, and finally generate a complete preservative traceability data set. Through the link data association module, the system uses a unique identifier to associate the production data and storage data of each batch of preservatives, and associates the dairy batch ID with the preservative batch ID to form a dairy component association table PCA, and establish a detailed traceability relationship between different batches. This module enhances the traceability and transparency of product quality, facilitates the rapid identification of the source of the problem when quality problems occur, helps reduce the recall costs caused by incomplete traceability, and improves production management efficiency.

[0052] (2) The system uses the preservative stability analysis module and the dairy product shelf life prediction module. The system establishes a component stability algorithm model and a preservative effect attenuation integral model through big data analysis, and calculates the preservative stability index S(t) and the comprehensive effect index E(t) in real time. The preservative stability analysis module can not only automatically calculate the stability changes of each batch of preservatives based on different storage environment data, but also further use the effect attenuation model to predict the preservative effect during the storage period of dairy products and generate a comprehensive effect index E(t). In the shelf life prediction module, the system uses the effect index E(t), raw material quality and process data to calculate the estimated shelf life value L, and compares it with the standard shelf life Tnorm. If the shelf life prediction value L is lower than 0.9, 0.8 and 0.7 times of Tnorm, the first-level warning information, second-level warning information and third-level warning information are generated respectively, reminding management personnel to pay attention to the stability of preservatives and changes in raw materials or processes in a timely manner. This function enables companies to dynamically predict the shelf life of dairy products, provide a scientific decision-making basis for the use of preservatives, improve the efficiency of preservative use, and reduce the risk of premature deterioration due to preservative failure.

[0053] (3) The system presents quality data and analysis results in an intuitive form through the data visualization and report generation module, forming a visual report, which further improves the company's monitoring and decision-making efficiency in quality management. This module generates the preservative stability report SIR, the shelf life prediction report LPR and the process monitoring report PMR, which respectively display the stability change curves of different batches of preservatives, the shelf life estimation of dairy products and related warning information, and monitor and analyze the changing trends of production process parameters. The preservative stability report SIR shows the changes in the stability index S(t) of different preservative batches. Through intuitive methods such as line graphs and color markings, managers can clearly see which batch of preservatives has decreased in stability and how much it has decreased. The shelf life prediction report LPR shows the shelf life prediction value and warning prompts of each batch of dairy products, helping management understand the quality status and potential risks of dairy products within the shelf life. The process monitoring report PMR sorts the production process parameters in time series and highlights the process parameters that deviate from the standard, so that management can quickly identify and adjust the production process. Through the graphical display of data and multi-level early warning mechanism, enterprises can achieve real-time quality monitoring, quickly respond to potential risks, and improve overall quality management efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 The present invention is a flowchart of a food quality and safety tracking management system based on big data. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] Example 1

[0057] See also Figure 1 The present invention provides a food quality and safety tracking management system based on big data. To achieve the above purpose, the present invention is implemented through the following technical scheme: including a parameter acquisition module, a link data association module, a preservative stability analysis module, a dairy product shelf life prediction module and a data visualization and report generation module;

[0058] The parameter acquisition module collects the quality data of the source nodes of dairy products through data acquisition equipment, and transmits the preservative traceability data to the central database, and performs preprocessing to obtain the preservative traceability data set;

[0059] The link data association module manages the preservative traceability data set in a unified manner through the big data platform, and uses unique identification of multi-table data to form a complete dairy ingredient association table PCA;

[0060] The preservative stability analysis module establishes a component stability algorithm model in the central database, calculates and outputs the preservative stability index S(t) based on the preservative traceability data set, and further constructs a preservative effect attenuation integral model based on the output preservative stability index S(t), and calculates and outputs the preservative comprehensive effect index E(t);

[0061] The dairy product shelf life prediction module analyzes the attenuation effect of the comprehensive effect index E(t) of the preservative to output the estimated shelf life value L, and then evaluates the estimated shelf life value L with the set normal shelf life Tnorm. Based on the evaluation results, a multi-level early warning mechanism is generated to indicate the stability of the preservative.

[0062] The data visualization and report generation module generates visualization reports by building a visualization platform, and uses visualization tools to generate corresponding graphical displays and analyses based on the visualization reports.

[0063] In this embodiment, the system builds a full-process, intelligent dairy product quality tracking and management system through the efficient collaboration of parameter acquisition module, link data association module, preservative stability analysis module, dairy product shelf life prediction module and data visualization and report generation module. The parameter acquisition module collects various quality data of dairy products in the production and storage process in real time through data acquisition equipment, generates a preservative traceability data set, and uniformly manages and integrates multi-table data through link data association module to form a complete component association table PCA, laying a solid foundation for data traceability. Based on these data, the preservative stability analysis module uses big data algorithm to calculate the stability index S(t) and comprehensive effect index E(t) of the preservative, and further analyzes the shelf life estimate L through the dairy product shelf life prediction module, and generates a multi-level warning after comparing with the standard shelf life Tnorm, indicating quality risks. Finally, the data visualization and report generation module presents the data analysis results in the form of a visual report, providing managers with graphical and intuitive quality analysis support. Compared with traditional quality inspection and management methods, this system has significantly improved the real-time nature of data acquisition, the comprehensiveness of data processing and the accuracy of quality prediction. Automated data collection and intelligent analysis reduce human interference and link ingredient quality with environmental conditions, providing multi-dimensional data support for quality management of preservatives and dairy products. At the same time, the system helps managers quickly identify and respond to potential quality issues through multi-level early warning mechanisms and visual analysis, significantly improving product quality stability and management efficiency.

[0064] Example 2

[0065] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the parameter acquisition module includes an acquisition unit and a pre-processing unit;

[0066] The collection unit collects the quality data of preservatives used in the production process of dairy products in real time by installing data collection equipment at the source node of dairy products;

[0067] Source nodes include production nodes and storage nodes, and quality data include production data and storage data;

[0068] The acquisition equipment installed at the production nodes includes Ph sensors, microbial online monitoring equipment, near-infrared spectrometers NIR, thermocouples and rotary encoders; the production data includes the Ph change value △PH of dairy products, microbial content Ws, fat content Zf, protein content Db, processing temperature Tj and stirring rate R;

[0069] The storage node is equipped with collection devices including temperature sensors, humidity sensors and oxygen concentration sensors; the stored data includes storage environment temperature T, storage environment humidity H and oxygen concentration O.

[0070] The preprocessing unit is used to transmit the collected quality data to the central database by connecting the sensor to the central database through the Internet of Things device, and preprocess the collected quality data in the central database. The preprocessing method includes data cleaning and standardization. After preprocessing, the quality data is timestamped and integrated according to the sensor collection time to obtain the preservative traceability data set, and stored in the central database in the order of timestamps;

[0071] The preservative traceability dataset includes production dataset and storage dataset;

[0072] The production data set includes the Ph change value △PH(t), microbial content Ws(t), fat content Zf(t) and protein content Db(t) of dairy products at time t;

[0073] The storage data set includes the storage environment temperature T(t), the storage environment humidity H(t), and the oxygen concentration O(t) at time t.

[0074] In this embodiment, the system realizes the efficient collection and integration of dairy production and storage data through the collaboration of the collection unit and the preprocessing unit of the parameter acquisition module. The collection unit installs collection equipment at the production node and the storage node respectively to obtain the quality data of the source node of dairy products in real time, covering the dynamic changes of product quality and environmental conditions. Through the Internet of Things device, the preprocessing unit transmits the data of each node to the central database, and performs data cleaning and standardization processing, and integrates the data according to the timestamp to form a preservative traceability data set. This module not only realizes the automation and real-time collection of data in the production and storage links, but also makes the data more accurate and consistent in subsequent analysis through standardized preprocessing.

[0075] Example 3

[0076] This embodiment is explained in Example 2. Please refer to Figure 1 ,Specifically: the link data association module includes a production management ,integration unit and a database association unit;

[0077] The production management integration unit tracks all preservatives used in dairy products through the big data platform, generates a preservative data sheet, records the supplier, generation date and preservative batch data, and then combines the data set stored in the central database with the preservative data sheet to generate a unique identifier and integrate to obtain the preservative batch ID;

[0078] The preservative batch ID consists of a 20-digit ID and a barcode. The four digits represent the supplier number, the six digits represent the production date, the four digits represent the batch number, the two digits represent the storage environment temperature, the two digits represent the storage environment humidity, and the two digits represent the oxygen concentration. For example, F001-241030-S001-20-50-01 is interpreted as: F001 represents the supplier number ID, 241030 represents the production date PD, S001 represents the batch number BN, 20 represents the storage environment temperature of 20 degrees, 50 represents the storage environment humidity controlled at 50 degrees, and 01 represents the oxygen concentration controlled at 1%;

[0079] During the process of adding preservatives, the preservative batch ID is entered into the production workshop management system, and automatically synchronized with the dairy product batch ID being produced into the central database, and the PCA of the associated dairy product ingredients is generated;

[0080] The dairy batch ID consists of a 16-digit ID, which includes a three-digit factory number, a four-digit production date, a three-digit production line number, and a three-digit product category; for example, A01-241030-B01-C01 is interpreted as: factory number A01, production date 241030, production line number B01, and product category C01.

[0081] The database association unit constructs query rules through the dairy product component association table PCA, corresponds each dairy product batch ID to a preservative batch ID, and then sets a traceability mark based on the production date of the dairy product and the production date of the preservative to track whether expired or high-risk batches of preservatives have been used;

[0082] The PCA association table of dairy ingredients includes dairy batch ID, preservative batch ID, and preservative production date;

[0083] The dairy product batch ID is used to indicate the production batch of the dairy product;

[0084] The preservative batch ID indicates the preservative batch used in each dairy product batch;

[0085] The preservative production date is used to confirm the production cycle and expiration date of the preservative to ensure that the preservative meets the requirements for use.

[0086] In this embodiment, the system manages preservative data uniformly on the big data platform through the production management integration unit, generates a unique identification preservative batch ID containing information such as supplier number, production date and storage environment, and automatically synchronizes the preservative batch ID and dairy product batch ID to the central database when the preservative is added, forming a dairy product component association table PCA. Through this table, the database association unit can quickly query the preservative batch information used for each batch of dairy products, and automatically set the traceability mark in combination with the production and use date, so as to timely discover and isolate product batches using expired or high-risk preservatives. Compared with the traditional method, the link data association module realizes the automatic association and efficient management of preservative batch data and dairy product batch data, significantly improving the accuracy and speed of data traceability. Through the unique identification and ingredient association table, enterprises can track the use of each batch of preservatives in real time and quickly lock potential risk batches. This module not only improves the traceability of product quality by recording and associating the use data of preservatives and dairy product batches throughout the process, but also provides strong data support for efficient management and quality control, and enhances the safety of raw materials and finished products.

[0087] Example 4

[0088] This embodiment is explained in Example 3, please refer to Figure 1 ,Specifically: the preservative stability analysis module includes a preservative stability analysis unit and a preservative effect attenuation unit;

[0089] The preservative stability analysis unit establishes a component stability algorithm model, extracts the stored data set and inputs it into the component stability algorithm model to output the preservative stability index S(t) of the preservative at time t;

[0090] The preservative stability index S(t) is calculated and output by the following ingredient stability algorithm model;

[0091]

[0092] In the formula, e represents the exponential function, α represents the basic coefficient of the influence of temperature on stability, β represents the nonlinear degree of temperature influence, which is used to adjust the nonlinear effect of storage environment temperature, γ represents the attenuation coefficient of humidity on stability, and t represents the unit time t.

[0093] The preservative effect attenuation unit constructs a preservative effect attenuation integral model based on the preservative stability index S(t) output by the component stability algorithm model, calculates and outputs the preservative comprehensive effect index E(t) at time t, and analyzes the attenuation effect of preservatives during the shelf life of dairy products.

[0094] The comprehensive effect index of preservatives E(t) is calculated and output by the following preservative effect attenuation integral model:

[0095]

[0096] Where κ represents the attenuation coefficient and dt represents the time-integrated variable.

[0097] In this embodiment, the method comprehensively evaluates the stability of preservatives under dairy storage conditions and their decay effects over time through the preservative stability analysis module through the combined action of the preservative stability analysis unit and the preservative effect decay unit. The preservative stability analysis unit uses the component stability algorithm model to input environmental data such as temperature and humidity into the model, dynamically calculates the preservative stability index S(t), and accurately reflects the impact of the environment on the stability of the preservative through parameters such as nonlinear temperature adjustment coefficient and humidity decay coefficient. Subsequently, the effect decay unit constructs a decay integral model based on the preservative stability index S(t) value, and calculates the preservative comprehensive effect index E(t) to quantify the effectiveness changes of preservatives during the shelf life of dairy products. Compared with traditional regular testing and empirical evaluation, this module accurately quantifies the actual effect changes of preservatives under different conditions through real-time dynamic calculation, significantly improving the scientificity and timeliness of preservative stability evaluation. Using the dynamic calculation of environmental data and algorithm models, the system can track the actual decay trend of preservatives in real time in different storage environments.

[0098] Example 5

[0099] This embodiment is explained in Example 4. Please refer to Figure 1 ,Specifically: the dairy product shelf life prediction module includes a ,prediction analysis unit and a quality fluctuation early warning unit;

[0100] The prediction and analysis unit calculates and outputs the estimated shelf life value L based on the obtained comprehensive effect index of preservatives E(t) and the production data set, and analyzes the actual shelf life of dairy products;

[0101] The estimated shelf life value L is calculated and output by the following algorithm formula;

[0102]

[0103] In the formula, Texp represents the expected shelf life, φ represents the influence coefficient of preservative effect on shelf life, ψ represents the influence coefficient of dairy raw material quality on shelf life, represents the quality of dairy raw materials, ω represents the influence coefficient of production process on shelf life, and (T(t)·R(t)) represents the dynamic production process parameters.

[0104] The quality fluctuation warning unit evaluates the shelf life based on the obtained shelf life estimate L and the normal shelf life Tnorm, analyzes the deviation between the predicted shelf life value and the expected standard value, and generates a multi-level warning mechanism based on the evaluation results. The specific evaluation contents are as follows;

[0105] When the estimated shelf life value L ≥ the normal shelf life Tnorm, it means that the current batch of dairy products is within the standard shelf life and the quality is stable, and no warning needs to be generated;

[0106] When the estimated shelf life value L is less than 0.9×normal shelf life Tnorm, a first-level warning message is generated, suggesting that the preservative batch and raw material quality be checked;

[0107] When the estimated shelf life value L is less than 0.8×normal shelf life Tnorm, a second-level warning message is generated, suggesting further confirmation of process parameters and storage environment;

[0108] When the estimated shelf life value L is less than 0.7×normal shelf life Tnorm, a third-level warning message is generated, which automatically triggers the quality control process, suspends the current dairy product shipment, and prompts relevant quality control personnel to investigate the problem.

[0109] In this embodiment, the dairy product shelf life prediction module of the system accurately predicts the shelf life of dairy products and monitors its quality fluctuations in real time through the dual functions of the prediction analysis unit and the quality fluctuation warning unit. The prediction analysis unit combines the comprehensive effect index of preservatives E(t), the production data set, the quality of dairy raw materials, process parameters and other factors to calculate the estimated shelf life value L of dairy products to reflect the dynamic changes of the actual shelf life. The quality fluctuation warning unit sets a multi-level warning mechanism based on the comparison between the estimated shelf life value L and the normal shelf life Tnorm. When the estimated shelf life value L is lower than different thresholds, such as 0.9, 0.8 and 0.7 times the normal shelf life Tnorm, the system automatically generates a first-level warning information, a second-level warning information or a third-level warning information, which prompts to check the preservative batch, raw material quality and process parameters respectively, and triggers the quality control process in severe cases. The dairy product shelf life prediction module greatly improves the accuracy and timeliness of the prediction. By dynamically analyzing the impact of the environment, raw materials and process parameters on the shelf life, enterprises can respond more flexibly to potential changes in product quality. The multi-level early warning mechanism can issue warnings at the early stages of quality risks, helping managers to promptly detect and deal with potential problems before products are shipped out, thereby reducing recall risks and ensuring the stable quality of dairy products.

[0110] Example 6

[0111] This embodiment is explained in Example 5. Please refer to Figure 1 Specifically: The visualization and report generation module generates visualization reports by building a visualization platform, combining the collected quality data set and the output preservative stability index S(t) and shelf life estimate L. The visualization reports include the preservative stability report SIR, the shelf life prediction report LPR and the process monitoring report PMR, and are graphically displayed through visualization tools;

[0112] The Preservative Stability Report SIR uses data analysis tools Python, Pandas and NumPy and visualization libraries Matplotlib, Plotly and D3.js to generate a line chart of the preservative stability index S(t) of multiple preservative batches, showing the preservative stability index S(t) curves of different preservative batches for easy and intuitive comparison. Color annotations are added to the chart to mark the mutation points of the preservative stability index S(t), making it easier for managers to find batches with reduced stability.

[0113] The shelf life prediction report LPR generates the first-level warning information, second-level warning information and third-level information through shelf life assessment, using a time series chart to display the dairy batches with warning information and the warning information prompt content;

[0114] The process monitoring report PMR uses visualization tools to sort the production status of dairy products in time series based on the production data set, and displays it through the use of line graphs. At the same time, standard processing parameters are set to highlight abnormal dairy processing batches in the production data set.

[0115] In this embodiment, the visualization and report generation module of the system uses a constructed visualization platform to graphically display core information such as quality data, preservative stability index S(t) and shelf life estimate L during the production and storage of dairy products, providing highly intuitive support for the company's quality monitoring and decision-making. The preservative stability report SIR generated by this module compares the stability curves of different batches of preservatives, highlights the stability mutation points with line graphs and color annotations, and helps managers quickly identify batches with reduced stability; the shelf life prediction report LPR displays warning information through a time series graph, indicating the warning level of the dairy batch and its specific prompts, intuitively reflecting the shelf life risk of the product; the process monitoring report PMR displays the production process data in the form of a line graph, and highlights the fluctuations in process parameters of abnormal batches, so as to facilitate timely optimization of the production process. Compared with traditional static reports, this module significantly improves the real-time and interactivity of the report, allowing managers to clearly see potential quality problems and their trend changes in data charts. Through integrated visualization and multi-level early warning prompts, this module not only greatly improves the efficiency of data interpretation, but also enables managers to more intuitively grasp the quality status of each production batch and make quick response decisions, thereby improving the intelligence level of product quality control and the accuracy of risk management.

[0116] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

Claims

1. A food quality and safety tracking management system based on big data, characterized by: It includes parameter collection module, link data association module, preservative stability analysis module, dairy product shelf life prediction module and data visualization and report generation module; The parameter acquisition module collects the quality data of the source node of dairy products through the data acquisition equipment, and transmits the preservative traceability data to the central database, and performs preprocessing to obtain the preservative traceability data set; The link data association module manages the preservative traceability data set in a unified manner through the big data platform, and uses unique identification of multi-table data to form a complete dairy ingredient association table PCA; The preservative stability analysis module establishes a component stability algorithm model in the central database, calculates and outputs the preservative stability index S(t) based on the preservative traceability data set, and further constructs a preservative effect attenuation integral model based on the output preservative stability index S(t), and calculates and outputs the preservative comprehensive effect index E(t); The dairy product shelf life prediction module outputs a shelf life estimate L by analyzing the attenuation effect of the preservative comprehensive effect index E(t), and then evaluates the shelf life by comparing the shelf life estimate L with the set normal shelf life Tnorm. Based on the evaluation results, a multi-level early warning mechanism is generated to indicate the stability of the preservative. The data visualization and report generation module generates visualization reports by building a visualization platform, and generates corresponding graphical presentations and analyses based on the visualization reports through visualization tools.

2. The food quality and safety tracking management system based on big data according to claim 1 is characterized by: The parameter acquisition module includes an acquisition unit and a preprocessing unit; The collection unit collects the quality data of preservatives used in the production process of dairy products in real time by installing data collection equipment at the source node of dairy products; The source node includes a production node and a storage node, and the quality data includes production data and storage data; The production node is equipped with collection equipment including Ph sensors, microbial online monitoring equipment, near infrared spectrometer NIR, thermocouples and rotary encoders; the production data includes Ph change value △PH of dairy products, microbial content Ws, fat content Zf, protein content Db, processing temperature Tj and stirring rate R; The storage node includes a temperature sensor, a humidity sensor and an oxygen concentration sensor through the installed collection equipment; the stored data includes the storage environment temperature T, the storage environment humidity H and the oxygen concentration O.

3. The food quality and safety tracking management system based on big data according to claim 2 is characterized by: The preprocessing unit is used to transmit the collected quality data to the central database by connecting the sensor to the central database through the Internet of Things device, and preprocess the collected quality data in the central database, wherein the preprocessing method includes data cleaning and standardization, and after preprocessing, the quality data is timestamped and integrated according to the sensor acquisition time, to obtain the preservative traceability data set, and store it in the central database in the order of timestamps; The preservative traceability dataset includes a production dataset and a storage dataset; The production data set includes the Ph change value △PH(t), microbial content Ws(t), fat content Zf(t) and protein content Db(t) of dairy products at time t respectively; The storage data set includes the storage environment temperature T(t), the storage environment humidity H(t) and the oxygen concentration O(t) at time t.

4. The food quality and safety tracking management system based on big data according to claim 3 is characterized by: The link data association module includes a production management integration unit and a database association unit; The production management integration unit tracks all preservatives used in dairy products through the big data platform, generates a preservative data table, records the supplier, generation date and preservative batch data, and then combines the data set stored in the central database with the preservative data table to generate a unique identifier, and integrates to obtain the preservative batch ID; The preservative batch ID is composed of a 20-digit ID and a barcode, wherein four digits represent the supplier number, six digits represent the production date, four digits represent the batch number, two digits represent the storage environment temperature, two digits represent the storage environment humidity, and two digits represent the oxygen concentration; During the process of adding preservatives, the preservative batch ID is entered into the production workshop management system, and automatically synchronized with the dairy product batch ID being produced into the central database, and the PCA of the associated dairy product ingredients is generated; The dairy product batch ID is composed of a 16-digit ID, which includes a three-digit factory number, a four-digit production date, a three-digit production line number and a three-digit product type.

5. The food quality and safety tracking management system based on big data according to claim 4 is characterized by: The database association unit constructs a query rule through the dairy product component association table PCA, corresponds each dairy product batch ID to a preservative batch ID, and then sets a traceability mark according to the production date of the dairy product and the production date of the preservative; The dairy product component association table PCA includes dairy product batch ID, preservative batch ID and preservative production date.

6. The food quality and safety tracking management system based on big data according to claim 3 is characterized by: The preservative stability analysis module includes a preservative stability analysis unit and a preservative effect attenuation unit; The preservative stability analysis unit establishes a component stability algorithm model, extracts a stored data set and inputs it into the component stability algorithm model to output a preservative stability index S(t) of the preservative at time t; The preservative stability index S(t) is calculated and output by the following component stability algorithm model; In the formula, e represents the exponential function, α represents the basic coefficient of the influence of temperature on stability, β represents the nonlinear degree of temperature influence, which is used to adjust the nonlinear effect of storage environment temperature, γ represents the attenuation coefficient of humidity on stability, and t represents the unit time t.

7. The food quality and safety tracking management system based on big data according to claim 6 is characterized by: The preservative effect attenuation unit constructs a preservative effect attenuation integral model through the preservative stability index S(t) output by the component stability algorithm model, calculates and outputs the preservative comprehensive effect index E(t) at time t, and analyzes the attenuation effect of preservatives during the shelf life of dairy products; The preservative comprehensive effect index E(t) is calculated and output by the following preservative effect attenuation integral model; Where κ represents the attenuation coefficient and dt represents the time-integrated variable.

8. The food quality and safety tracking management system based on big data according to claim 1 is characterized by: The dairy product shelf life prediction module includes a prediction analysis unit and a quality fluctuation early warning unit; The prediction and analysis unit calculates and outputs a shelf life estimate L based on the obtained preservative comprehensive effect index E(t) in combination with the production data set, and analyzes the actual shelf life of the dairy product; The estimated shelf life value L is calculated and outputted by the following algorithm formula; In the formula, Texp represents the expected shelf life, φ represents the influence coefficient of preservative effect on shelf life, ψ represents the influence coefficient of dairy raw material quality on shelf life, represents the quality of dairy raw materials, ω represents the influence coefficient of production process on shelf life, and (T(t)·R(t)) represents the dynamic production process parameters.

9. A food quality and safety tracking management system based on big data according to claim 8, characterized in that: The quality fluctuation warning unit evaluates the shelf life and analyzes the deviation between the predicted shelf life value and the expected standard value based on the obtained estimated shelf life value L and the normal shelf life Tnorm, and generates a multi-level warning mechanism based on the evaluation results. The specific evaluation contents are as follows; When the estimated shelf life value L ≥ the normal shelf life Tnorm, it means that the current batch of dairy products is within the standard shelf life and the quality is stable, and no warning needs to be generated; When the estimated shelf life value L is less than 0.9×normal shelf life Tnorm, a first-level warning message is generated, suggesting that the preservative batch and raw material quality be checked; When the estimated shelf life value L is less than 0.8×normal shelf life Tnorm, a second-level warning message is generated, suggesting further confirmation of process parameters and storage environment; When the estimated shelf life value L is less than 0.7×normal shelf life Tnorm, a third-level warning message is generated, which automatically triggers the quality control process, suspends the current dairy product shipment, and prompts relevant quality control personnel to investigate the problem.

10. The food quality and safety tracking management system based on big data according to claim 9, characterized in that: The visualization and report generation module generates a visualization report by constructing a visualization platform, combining the collected quality data set and the output preservative stability index S(t) and shelf life estimate L. The visualization report includes a preservative stability report SIR, a shelf life prediction report LPR and a process monitoring report PMR, and is graphically displayed through a visualization tool; The preservative stability report SIR generates a line chart using data analysis tools and a visualization library to analyze the preservative stability index S(t) of multiple preservative batches, showing the preservative stability index S(t) curves of different preservative batches, and adds color annotations to the chart to mark the mutation points of the preservative stability index S(t); The shelf life prediction report LPR generates the first-level warning information, second-level warning information and third-level information through the shelf life evaluation, and uses a time series graph to display the dairy product batches with warning information and the warning information prompt content; The process monitoring report PMR uses a visualization tool to sort the production status of dairy products in time series based on the production data set, and displays it by using a line graph. At the same time, standard processing parameters are set to highlight abnormal dairy processing batches in the production data set.

Images