A digital visual meat quality food safety production monitoring management system

Abstract

The present application relates to the technical field of meat food safety production monitoring and management, in particular to a digital visual meat food safety production monitoring and management system. The system comprises a data acquisition and binding unit, an environmental stress modeling unit, a microbial growth prediction unit, a dynamic risk assessment unit, and a risk early warning and decision support unit. The data acquisition and binding unit is used to obtain environment time series data and static physicochemical data associated with the unique identity code of each batch of meat products. The environmental stress modeling unit is used to calculate the comprehensive environmental stress factor. The microbial growth prediction unit is used to calculate the predicted total amount of microorganisms. The dynamic risk assessment unit is used to calculate the dynamic safety risk index. The risk early warning and decision support unit is used to generate a risk level signal and solve the predictive remaining shelf life based on the preset critical risk threshold to support scheduling decisions. The system lays a solid data foundation for dynamic, continuous, and forward-looking microbial risk assessment of the entire production chain, and achieves significant technological progress.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of meat food safety production monitoring and management, in particular to a digital visual meat food safety production monitoring and management system. BACKGROUND

[0002] During the production and circulation of meat food, the growth of microorganisms is a key factor affecting the quality and safety of meat food; traditional food safety monitoring methods mainly rely on sampling detection at key control points, such as measuring the total number of colonies by microbial culture method;

[0003] Such traditional methods have significant limitations: long detection period, results have lagging nature, cannot provide real-time risk warning and decision support for the production process; sampling belongs to discrete static monitoring, which is difficult to capture and evaluate the dynamic and cumulative effects of environmental factors on microbial growth in the entire production chain; for example, a short interruption of the cold chain can lead to rapid proliferation of microorganisms, and this cumulative risk may not be accurately reflected in subsequent single-point sampling;

[0004] Therefore, the prior art urgently needs a monitoring and management method that can dynamically, continuously and prospectively evaluate the risk of meat food production throughout the chain, to solve the technical problem that traditional discrete and instantaneous detection indicators cannot effectively evaluate the dynamic and cumulative risk of food, from passive response to active prediction, and improve the scientificity and forward-looking nature of food safety management.

[0005] The above information disclosed in the above background section is only used to strengthen the understanding of the background of the present disclosure, and therefore it can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0006] To solve the above technical problems, the present application discloses a digital visual meat food safety production monitoring and management system, in particular, the technical scheme of the present application comprises:

[0007] A data acquisition and binding unit is used to obtain environment time series data and static physicochemical data associated with the unique identity code of each batch of meat products;

[0008] An environmental stress modeling unit is used to calculate a comprehensive environmental stress factor based on the environment time series data and the static physicochemical data through a preset mapping function;

[0009] A microbial growth prediction unit is used to combine the comprehensive environmental stress factor, the initial number of colonies contained in the static physicochemical data, and the preset maximum specific growth rate of the target strain, and calculate the predicted total amount of microorganisms by solving a microbial growth prediction model;

[0010] a dynamic risk assessment unit configured to calculate a dynamic safety risk index based on the predicted total amount of microorganisms and the comprehensive environmental stress factor;

[0011] a risk warning and decision support unit configured to compare and analyze the dynamic safety risk index with preset grading warning thresholds, generate a risk level signal, and solve a predictive remaining shelf life based on a preset critical risk threshold to support scheduling decisions.

[0012] Preferably, the environmental time-series data is temperature and humidity collected in real time by Internet of Things sensors deployed at key control points of slaughtering, cutting, warehousing, and logistics; and the static physicochemical data is initial colony count, initial pH value, and initial water activity detected upon entering the factory.

[0013] Preferably, the comprehensive environmental stress factor is obtained by multiplying a plurality of independent stress sub-factors corresponding to temperature, humidity, initial pH value, and initial water activity; wherein the function form and internal parameters of each independent stress sub-factor are obtained by regression analysis or machine learning training on historical batch real environment data and corresponding microorganism detection results.

[0014] Preferably, the microorganism growth prediction model accumulates the environmental impact in the production process by introducing an integral form; and the predicted total amount of microorganisms is obtained by multiplying the initial colony count and the exponential function value of the cumulative calculation result.

[0015] Preferably, the dynamic risk assessment unit first calculates the instantaneous growth rate of microorganisms based on the predicted total amount of microorganisms, the maximum specific growth rate of the target strain, and the comprehensive environmental stress factor.

[0016] Preferably, the dynamic risk assessment unit further performs the following steps to calculate the dynamic safety risk index:

[0017] determining a stock risk term based on the predicted total amount of microorganisms and a preset national food safety standard total plate count limit;

[0018] determining an incremental risk term based on the instantaneous growth rate of microorganisms, a preset characteristic time constant, and the national food safety standard total plate count limit;

[0019] performing weighted summation on the stock risk term and the incremental risk term in combination with a preset weight factor to generate the dynamic safety risk index.

[0020] Preferably, the risk warning and decision support unit generates the risk level signal as follows:

[0021] if the dynamic safety risk index is less than or equal to a preset first warning threshold, a normal signal is generated;

[0022] If the dynamic security risk index is greater than the first early warning threshold and less than or equal to the preset second early warning threshold, a concern signal is generated;

[0023] If the dynamic security risk index is greater than the second early warning threshold, an early warning signal is generated.

[0024] Preferably, when the generated risk level signal is an early warning signal, the risk early warning and decision support unit is further used to backtrack the historical values of the comprehensive environmental stress factor, locate the time period when the comprehensive environmental stress factor is abnormally high, and trace the key production link causing risk accumulation.

[0025] Compared with the prior art, the present application has the following beneficial effects:

[0026] 1、The system creates a unique identity code for each batch of products, integrates the environmental time series data such as temperature and humidity collected in real time by Internet of Things sensors, and the initial colony number, pH value and other static physicochemical data detected when entering the factory, and builds a complete digital file throughout the production process. This overcomes the defects of data isolation and inability to trace in traditional methods, and through the integration of discrete data points into continuous process records, it lays a solid data foundation for dynamic, continuous and forward-looking microbial risk assessment of the entire production chain, and realizes significant technological progress.

[0027] 2、The environmental stress modeling unit set up in the system converts multiple independent environmental parameters such as temperature, humidity, initial pH value and water activity into a comprehensive environmental stress factor through a mapping function generated by training historical data and microbial detection results. This design scientifically quantifies the synergistic effect of multiple factors on microbial growth, surpassing the limitations of traditional single variable monitoring, and endowing the model with self-adaptive and self-optimizing capabilities, enabling it to accurately reflect the real environmental cumulative impact under specific processes, providing reliable input for subsequent accurate prediction.

[0028] 3、The system generates a dynamic security risk index by weighting and summing the stock risk, comparison of the current predicted total number of colonies with the national standard limit value, and the incremental risk based on the deterioration trend of instantaneous growth rate. This evaluation paradigm that comprehensively considers the current situation and trend can provide very early warning for batches with high risk potential that have acceptable current microbial levels but rapid growth, and significantly advance the window period of risk management. It fundamentally solves the pain points of traditional sampling inspection that cannot capture the process history impact and risk dynamic evolution trend, and realizes the transition from passive response to active prediction.

[0029] 4、The system converts the risk assessment results into specific management instructions. The risk early warning and decision support unit can not only generate graded early warning signals based on the comparison of the dynamic safety risk index and the preset threshold to achieve the refinement of management, but also can automatically backtrack the historical values of the comprehensive environmental stress factors to accurately locate the key production link causing the risk accumulation when a high-level early warning is triggered. This function provides a powerful diagnostic tool, provides direct data basis for process optimization, realizes the improvement from processing problem products to eliminating the source of problems, and provides scientific decision support for supply chain management through the predicted remaining shelf life. BRIEF DESCRIPTION OF DRAWINGS

[0030] The application will be further explained below in conjunction with the accompanying drawings and embodiments:

[0031] Figure 1 is a system structure diagram of the application. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical scheme and advantages of the application more clear and obvious, the application will be further described in detail below in conjunction with specific embodiments. Embodiment 1:

[0033] Please refer to Figure 1 A digital visual meat food safety production monitoring management system, comprising:

[0034] A data acquisition and binding unit for acquiring environmental time series data and static physicochemical data associated with the unique identity code of each batch of meat products;

[0035] An environmental stress modeling unit for calculating a comprehensive environmental stress factor based on the environmental time series data and the static physicochemical data through a preset mapping function;

[0036] A microbial growth prediction unit for combining the comprehensive environmental stress factor, the initial bacterial colony number contained in the static physicochemical data, and the preset maximum specific growth rate of the target bacteria to calculate the predicted total amount of microorganisms by solving a microbial growth prediction model;

[0037] A dynamic risk assessment unit for calculating a dynamic safety risk index based on the predicted total amount of microorganisms and the comprehensive environmental stress factor;

[0038] A risk early warning and decision support unit for comparing and analyzing the dynamic safety risk index with a preset graded early warning threshold to generate a risk level signal, and solving a predicted remaining shelf life based on a preset critical risk threshold to support scheduling decisions;

[0039] This embodiment provides a digital visualization meat food safety production monitoring and management system. The system aims to solve the problem that traditional discrete and instantaneous detection indicators cannot effectively assess the dynamic cumulative risks of food throughout the entire production chain. By constructing a digital twin model, the system can predict the microbial growth status in real time and continuously, and conduct forward-looking risk assessment and decision support.

[0040] The complete technical loop of this system is accomplished through the collaborative work of the following units:

[0041] The data acquisition and binding unit establishes a comprehensive, time-series digital file for each independent production batch; in this embodiment, when a batch of meat enters the production process, the system assigns it a unique identification code (…). This unit collects real-time data on the batch through IoT sensors deployed at key control points such as slaughtering, processing, warehousing, and logistics. Environmental time-series data that are uniquely correlated mainly includes temperature. With humidity Meanwhile, static physicochemical data of this batch are obtained through an interface or tested upon arrival at the factory, mainly the initial total bacterial count. Initial pH value and initial water activity In this model, to simplify calculations, the initial pH value and water activity are considered as inherent batch properties that remain constant throughout the monitoring period; in future model iterations, they can also be treated as dynamic variables that change with microbial growth.

[0042] All of this dynamic and static data is bound to this batch. After data cleaning and normalization, the data cleaning steps include validating the effective range of sensor data, for example, limiting the temperature range to between -20℃ and 50℃, and smoothing outliers or missing values ​​outside the range by using methods such as nearest neighbor interpolation or moving average filling, to ensure the robustness of the model and form the data foundation required for subsequent modeling.

[0043] The environmental stress modeling unit transforms multiple independent environmental parameters into a single index that comprehensively reflects their overall impact on microbial growth; to further clarify, this embodiment introduces a comprehensive environmental stress factor. To quantify the comprehensive impact of environmental parameters, this unit uses a pre-defined mapping function to calculate... The specific calculation method is as follows:

[0044] ;

[0045] in, : Comprehensive environmental stress factor, dimensionless, calculated by this unit based on each input parameter;

[0046] : Independent stress sub-factor function corresponding to temperature, humidity, pH value and water activity respectively; Its function form and internal undetermined parameters are obtained by regression analysis or machine learning training on a large number of historical batches of real environment data and corresponding microbial detection results;

[0047] : Real-time environmental parameters, is the real-time temperature, is the real-time humidity, provided by the data acquisition and binding unit;

[0048] : Batch-specific static physicochemical parameters, is the initial pH value, is the initial water activity, provided by the data acquisition and binding unit;

[0049] Based on the recognized theoretical model in the field of food microbiology, the independent influence of each environmental factor on the growth rate of microorganisms can be multiplied and superimposed; this embodiment will this theoretical model, thereby effectively combining the real-time changing dynamic parameters ( ) and batch-specific static parameters ( ) The output value can more accurately reflect the comprehensive promotion or inhibition effect of the real environment on microbial growth;

[0050] Microbial growth prediction unit, based on the quantification results of environmental influence, a microbial growth prediction model that can reflect the cumulative effect is established; this model is an improvement of the classic exponential growth model, by introducing the integral form to reflect the continuous influence of the process history; in order to calculate the predicted total amount of microorganisms at any time , the unit adopts the following formula:

[0051] ;

[0052] Wherein, : predicted total amount of microorganisms at time, unit: CFU / g, calculated by this unit;

[0053] : Initial total number of colonies, unit: CFU / g, provided by the data acquisition and binding unit;

[0054] : Maximum specific growth rate of target species under ideal conditions, unit: ​, which is obtained from a public database according to the meat category as a key parameter representing the growth rate of the entire microbial community; in practical applications, this parameter can be selected as the growth rate of the target dominant spoilage bacteria, or an equivalent rate calibrated according to historical data to represent the overall proliferation trend of the entire microbial community;

[0055] : historical time : comprehensive environmental stress factor provided by the environmental stress modeling unit;

[0056] The internal logic is that the system continuously updates the predicted value of by performing real-time numerical integration on all historical time ; this design allows the impact of any transient environmental fluctuations in the production chain to be accurately accumulated into the final result, solving the problem that traditional detection methods cannot evaluate historical impacts;

[0057] The dynamic risk assessment unit realizes forward-looking risk warning and comprehensive assessment of the safety state of the product, considering not only the current total amount of microorganisms but also their growth trend; for this purpose, the embodiment defines a dynamic safety risk index ; the technical concept of the embodiment is that a rapidly deteriorating product has a higher instantaneous risk than a slowly deteriorating product; the calculation formula is as follows:

[0058]

[0059] Before calculating , the unit needs to calculate the instantaneous growth rate of microorganisms , which can be directly derived from the calculation formula of :

[0060] ;

[0061] Wherein, : dynamic safety risk index, dimensionless, calculated by the unit;

[0062] : current microbial predicted total amount, the calculation result of the microbial growth prediction unit;

[0063] : instantaneous growth rate of microorganisms, calculated by the unit according to and ;

[0064] : total amount of colonies, statutory limit, preset according to national food safety standards;

[0065] Weighting factors, dimensionless, are predetermined based on expert experience. Their values ​​are selected to balance the contribution of existing and incremental risks in the final risk assessment. For example, when monitoring meat products with rapid spoilage characteristics, the weighting factor representing incremental risk can be appropriately increased. The proportion of;

[0066] The characteristic time constant is a preset value (e.g., 1 hour) used for dimensional uniformity.

[0067] By reflecting the current state of stock risk ( The sum is weighted and summed with the incremental risk reflecting future trends, i.e., the growth rate term. It can identify potential security risks earlier and more accurately;

[0068] The risk warning and decision support unit transforms risk assessment results into specific, actionable management instructions; this unit is based on a risk index. The value is compared and analyzed with the preset graded early warning thresholds to generate risk signals of normal, attention, or warning levels, which are then pushed to management personnel; graded early warning thresholds, such as attention thresholds... and warning threshold Its value is determined based on statistical analysis of a historical security incident database; specifically, it is determined by analyzing historical batches. The correlation between the trajectory and whether the final standard is exceeded is determined using methods such as receiver operating characteristic curve analysis, selecting a point that achieves a preset balance between specificity and sensitivity. The value is used as a threshold;

[0069] In addition, to provide proactive decision support, this unit also sets a critical risk threshold. Solve the equation forward The time in This allows for the calculation of the predicted remaining shelf life of the batch. When solving this equation, the system defaults to using standard, preset warehousing or logistics environment parameters, such as a constant refrigerated temperature of 4°C and humidity of 75%, as environmental data input for future time periods to ensure the consistency and comparability of the predictions. Values ​​can be used to guide the supply chain in implementing dynamic first-in-first-out (FIFO) strategies and prioritizing scheduling. Shortest batch sizes enable refined inventory management;

[0070] The embodiment builds a complete technical closed loop from data acquisition to decision support through the cooperation of the above-mentioned units; it overcomes the limitations of traditional food safety monitoring relying on discrete and static detection indicators, realizes dynamic, continuous and cumulative evaluation of microbial risks of meat food in the whole production chain through the establishment of a digital twin model; this not only improves the timeliness and accuracy of risk early warning, but also provides accurate data decision basis for supply chain management through functions such as remaining shelf life prediction, and improves the scientificity and forward-looking of food safety management. Embodiment 2:

[0071] The environmental time series data is the temperature and humidity collected in real time by the Internet of Things sensors deployed at key control points of slaughtering, cutting, warehousing and logistics; the static physicochemical data is the initial number of colonies, initial pH value and initial water activity detected when entering the factory;

[0072] The embodiment is a specific implementation mode of the data acquisition and binding unit based on embodiment 1; in order to ensure the accuracy and relevance of the model input, the embodiment optimizes and limits the source and type of data;

[0073] The environmental time series data is specifically defined as the temperature ( ) and humidity ( ) collected in real time by the Internet of Things sensors deployed at key control points such as slaughtering, cutting, warehousing and logistics;

[0074] The static physicochemical data is specifically defined as the initial number of colonies ( ), initial pH value ( ) and initial water activity ( ) detected when the meat batch enters the factory;

[0075] By limiting the data source to these core physical and chemical parameters most directly related to the growth of meat microorganisms, the scheme ensures high fidelity and strong correlation of the input data; compared with the use of generalized environmental data, this targeted data acquisition strategy greatly improves the accuracy of subsequent environmental stress modeling and microbial growth prediction, laying a solid data foundation for the reliability of the entire system. Embodiment 3:

[0076] The comprehensive environmental stress factor is obtained by multiplying a plurality of independent stress sub-factors corresponding to temperature, humidity, initial pH value and initial water activity; wherein the function form and internal parameters of each independent stress sub-factor are obtained by regression analysis or machine learning training on the real environmental data of historical batches and the corresponding microbial detection results;

[0077] The embodiment is a specific implementation mode of the environmental stress modeling unit based on embodiment 1, which calculates the comprehensive environmental stress factor ( A preferred method;

[0078] In this embodiment, It was explicitly designed to be obtained by multiplying four independent stress sub-factors corresponding to temperature, humidity, initial pH, and initial water activity, respectively; each independent stress sub-factor, for example, a bell-shaped function describing the effect of temperature. Its specific form can be represented by a quadratic polynomial model, such as:

[0079] ;

[0080] in, and These are the minimum and maximum temperature thresholds for microbial growth, respectively. Their functional form and internal parameters are not pre-fixed empirical values, but rather obtained through analysis of datasets containing multiple historical batches. Each dataset contains a set of historical environmental parameters, for example, at a constant temperature. Below are the corresponding measured final values ​​of microorganisms. By employing mathematical methods such as least squares regression analysis, the undetermined parameters in each sub-factor function are calibrated to ensure that the model's predictions are consistent with the actual results. Minimize the error;

[0081] The product form used in this embodiment is an effective and widely accepted model that assumes the influence of each environmental factor is relatively independent. In scenarios with higher accuracy requirements, interaction terms can be introduced into the model, for example... To more accurately characterize the interaction between specific factors;

[0082] In this preferred method, the product form achieves a good balance between computational efficiency and model accuracy, conforms to basic microbiological principles, and more realistically reflects the synergistic effects of multiple environmental factors, making it more accurate than a simple linear superposition model. By using machine learning to determine the sub-factor functions, the model acquires adaptive and self-optimizing capabilities, learning unique influence patterns under specific production processes and environments from historical data, thereby generating a highly customized and more accurate model. This improved the accuracy of the entire prediction system;

[0083] Other independent stress sub-factors can also be modeled using specific functional forms; for example, the water activity stress sub-factor. A monotonically increasing function can be used, such as when Below the minimum water activity for microbial growth hour higher than Time can be represented as pH stress factor A bell-shaped function similar to temperature can also be used for modeling; the form and parameters of these functions are determined by regression analysis or machine learning training on historical data to achieve model adaptation and self-optimization.

[0084] Furthermore, the model exhibits good physical boundary behavior; when the ambient temperature... At the lowest growth temperature or highest temperature hour, The value is 0, at which point the instantaneous growth rate is 0, and the total number of microorganisms stops growing, which is consistent with common biological knowledge; if the temperature exceeds this range... When the value becomes negative, the model will predict a decrease in the total amount of microorganisms, which can be used to simulate the lethal effects of low or high temperatures on microorganisms. When the system performs predictive remaining shelf life calculations, it will handle negative growth cases specially, such as counting them as an extension or indefinite extension of shelf life, to ensure the rationality of the output results. Example 4:

[0085] The microbial growth prediction model incorporates an integral form to cumulatively calculate the environmental impact of the production process; the predicted total microbial count is obtained by multiplying the initial colony count by the exponential function value of the cumulative calculation result.

[0086] This embodiment is based on Embodiment 1, and elaborates on the specific microbial growth prediction model used in the microbial growth prediction unit;

[0087] This model incorporates an integral form to cumulatively calculate the environmental impact at all historical moments during the production process; as shown in the aforementioned formula. As shown, the final predicted total microbial population ( ) is determined by the initial colony count ( This is obtained by multiplying the value of the exponential function, which represents the cumulative effect of the entire process, by the value of the exponential function.

[0088] The introduction of an integral form is a key innovation of this scheme; it enables the precise quantification of the irreversible accumulation of the impact of adverse environmental factors in the process history, ensuring that any short-term but drastic environmental fluctuations, such as the impact of cold chain interruptions, can be faithfully recorded and accumulated, and will not be ignored when the environment returns to normal. This mechanism solves the pain point that traditional sampling inspection cannot capture the impact of the process history, making the prediction of the final product quality more reliable and in line with the facts.

[0089] It should be noted that this model is an exponential growth model, which can accurately characterize the dynamics of microorganisms during the exponential growth phase, and is crucial for early risk warning. This model does not include parameters describing the growth plateau phase because, in food safety management, when microorganisms approach the plateau phase, the product has usually already exceeded the limits and lost its edible value. If the study needs to cover the entire growth curve, the stress factors in this model can also be included. Thought and Classics or The models are combined. Example 5:

[0090] The dynamic risk assessment unit calculates the instantaneous growth rate of microorganisms based on the predicted total microbial population, the maximum specific growth rate of the target species, and comprehensive environmental stress factors.

[0091] The dynamic risk assessment unit further performs the following steps to calculate the dynamic security risk index:

[0092] Based on the predicted total microbial count and the preset national food safety standard total bacterial count limit, existing risk items are identified;

[0093] Incremental risk items are determined based on the instantaneous growth rate of microorganisms, the preset characteristic time constant, and the total colony limit of national food safety standards.

[0094] By combining preset weighting factors, the existing risk items and the incremental risk items are weighted and summed to generate a dynamic safety risk index;

[0095] This embodiment, based on Embodiment 1, explains how the dynamic risk assessment unit calculates the dynamic security risk index (…). This is a more forward-looking risk assessment process;

[0096] The dynamic risk assessment unit is based on the predicted total microbial population ( ), the maximum specific growth rate of the target bacterial species ( ) and comprehensive environmental stress factors ( ), through formula The instantaneous growth rate of the microorganisms was calculated;

[0097] This unit further performs the following steps to calculate... :

[0098] Identifying existing risk items: based on predicted total microbial count ( ) and the preset national food safety standard total bacterial count limit ( ), calculate the ratio and multiply by its weighting factor This item reflects how close the product's current pollution level is to the safety limit.

[0099] Identifying incremental risk items: based on the instantaneous growth rate of microorganisms ( ), preset characteristic time constant ( )and ,calculate and multiply by its weighting factor This item reflects the deterioration trend or speed of the product's safety status;

[0100] Weighted summation: combining preset weight factors and The existing and new risk items are weighted and summed to generate the final dynamic safety risk index. ;

[0101] This comprehensive assessment paradigm, which considers both existing and new risks, not only evaluates the current state of pollution but, more importantly, assesses its rate of deterioration. This enables the system to provide early warnings for high-risk potential batches that currently have acceptable levels of microorganisms but exhibit extremely rapid growth trends, thereby advancing the window of opportunity for risk management and achieving a shift from passive response to proactive prediction. Example 6:

[0102] The process by which the risk warning and decision support unit generates risk level signals is as follows:

[0103] If the dynamic safety risk index is less than or equal to the preset first warning threshold, a regular signal is generated;

[0104] If the dynamic safety risk index is greater than the first warning threshold and less than or equal to the preset second warning threshold, a concern signal is generated.

[0105] If the dynamic safety risk index is greater than the second warning threshold, a warning signal will be generated;

[0106] When the generated risk level signal is an early warning signal, the risk early warning and decision support unit is also used to trace back the historical values ​​of the comprehensive environmental stress factor, locate the period when the comprehensive environmental stress factor is abnormally high, and trace the key production links that cause the risk accumulation.

[0107] Based on Example 1, this embodiment further refines the functions of the risk warning and decision support unit, demonstrating how it transforms risk indices into specific management actions.

[0108] The process by which this unit generates risk level signals is as follows:

[0109] If the dynamic security risk index ( () less than or equal to the preset first warning threshold If so, a normal signal is generated, indicating that the state is normal;

[0110] like Greater than And less than or equal to the preset second warning threshold If so, a warning signal will be generated to alert management personnel.

[0111] like Greater than if so, a pre-warning signal is generated, triggering a high-level alarm;

[0112] When the generated risk level signal is a pre-warning signal, the unit is also activated to perform a key traceability function: it automatically traces back the historical values of the integrated environmental stressor (IES) ) and precisely locates the time interval and the corresponding production stage that contributed most to the risk score by analyzing the time series curve of the IES .

[0113] It should be noted that the above examples are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. A digital visualized meat quality food safety production monitoring management system, characterized in that, The method comprises the following steps: a data acquisition and binding unit is used to acquire environmental time series data and static physicochemical data associated with the unique identity code of each batch of meat products; an environmental stress modeling unit is used to calculate a comprehensive environmental stress factor based on the environmental time series data and the static physicochemical data through a preset mapping function; the comprehensive environmental stress factor is obtained by multiplying a plurality of independent stress sub-factors corresponding to temperature, humidity, initial pH value and initial water activity; the function form and internal parameters of each independent stress sub-factor are obtained by regression analysis or machine learning training on the real environmental data of historical batches and the corresponding microbial detection results; a microbial growth prediction unit is used to combine the comprehensive environmental stress factor, the initial bacterial colony number contained in the static physicochemical data and the preset maximum specific growth rate of the target strain, and calculate the predicted total amount of microorganisms by solving a microbial growth prediction model; the microbial growth prediction model calculates the cumulative effect of the environment during production by introducing an integral form; the predicted total amount of microorganisms is obtained by multiplying the initial bacterial colony number and the exponential function value of the cumulative calculation result; a dynamic risk assessment unit is used to calculate a dynamic safety risk index based on the predicted total amount of microorganisms and the comprehensive environmental stress factor; the dynamic risk assessment unit first calculates the instantaneous growth rate of microorganisms based on the predicted total amount of microorganisms, the maximum specific growth rate of the target strain and the comprehensive environmental stress factor; a risk warning and decision support unit is used to compare and analyze the dynamic safety risk index with preset graded warning thresholds, generate a risk level signal, and solve a predictive remaining shelf life based on a preset critical risk threshold to support scheduling decisions; The dynamic risk assessment unit further performs the following steps to calculate the dynamic safety risk index: determine the inventory risk term based on the predicted total amount of microorganisms and the preset national food safety standard total plate count limit; determine the incremental risk term based on the instantaneous growth rate of microorganisms, the preset characteristic time constant and the national food safety standard total plate count limit; combine the inventory risk term and the incremental risk term by weighted summation with a preset weight factor to generate the dynamic safety risk index.

2. The digital visualized meat quality food safety production monitoring management system according to claim 1, characterized in that, The environmental time series data is the temperature and humidity collected in real time by Internet of Things sensors deployed at key control points of slaughtering, cutting, warehousing and logistics; the static physicochemical data is the initial bacterial colony number, initial pH value and initial water activity detected when entering the factory.

3. The digital visualized meat food safety production monitoring and management system according to claim 1, characterized in that, The risk warning and decision support unit generates the risk level signal as follows: if the dynamic safety risk index is less than or equal to the preset first warning threshold, a normal signal is generated; if the dynamic safety risk index is greater than the first warning threshold and less than or equal to the preset second warning threshold, an attention signal is generated; if the dynamic safety risk index is greater than the second warning threshold, a warning signal is generated.

4. The digital visualized meat food safety production monitoring and management system according to claim 1, characterized in that, When the generated risk level signal is the warning signal, the risk warning and decision support unit is further used to backtrack the historical values of the comprehensive environmental stress factor, locate the time period when the comprehensive environmental stress factor is abnormally high, and trace the key production link that causes risk accumulation.

Images