A food supply chain security monitoring and early warning system based on big data analysis

Abstract

The invention discloses a food supply chain safety monitoring and early warning system based on big data analysis, comprising: a data acquisition module, configured to capture preset data of each link of the food supply chain in a preset area; a data analysis module, configured to analyze the preset data to obtain a predicted food demand; a data monitoring module, configured to screen historical data matching the predicted demand, and screen historical data with the most variable overlap, set the overlapping variables in the preset data as first variables, set the remaining variables as second variables, monitor the second variables, and set early warning indicators according to the second variables and historical data; an early warning release module, configured to release early warning information to an early warning center; an emergency response module, configured to generate a corresponding emergency plan according to the early warning information; the invention adjusts the preset data by comparing the predicted food demand with the historical data, so that the terminal supply of food meets the predicted demand.

Description

Technical Field

[0001] The present invention belongs to the field of food supply chain, and in particular relates to a food supply chain safety monitoring and early warning system based on big data analysis. Background Art

[0002] The food supply chain includes all links from food production, processing, storage, circulation to consumption. There are hundreds of millions of farmers and tens of thousands of small and medium-sized agricultural enterprises in my country's food supply chain. With the development of urbanization and the gradual formation of the "North Grain South Transportation" production and marketing pattern, the food supply chain continues to extend, and the number of key nodes continues to increase. The various links are interconnected and influence each other. The pressure on the middle links is increasing, and the overall resilience of the supply chain is easily affected.

[0003] If it is necessary to accurately control the data indicators of each link in the food supply chain, it will be extremely difficult. Summary of the invention

[0004] Purpose of the invention: In order to overcome the above shortcomings, the purpose of the present invention is to provide a food supply chain security monitoring and early warning system based on big data analysis, which can improve the overall resilience of the food supply chain and reduce the data indicators of the intermediate links that need to be monitored.

[0005] Technical solution: A food supply chain security monitoring and early warning system based on big data analysis, including:

[0006] A data collection module is configured to capture preset data of each link of the food supply chain in a preset area and integrate the preset data into a database, wherein the preset data is dynamic data, the preset data includes a number of variables, and the preset data is updated at a preset frequency;

[0007] A data analysis module is configured to process and analyze the preset data, and obtain a predicted demand for future food by performing trend analysis on historical data;

[0008] A data monitoring module is configured to screen historical data that matches the predicted demand, and screen historical data with the most overlapping variables, set the overlapping variables in the preset data as first variables, set the remaining variables as second variables, monitor the second variables, and set early warning indicators according to the second variables and historical data;

[0009] The early warning release module is configured to release corresponding food supply chain security early warning information to the early warning center based on monitoring conditions and early warning indicators;

[0010] An emergency response module is configured to generate a corresponding emergency plan according to the warning information;

[0011] By comparing historical data with overlapping variables, the need to monitor the second variable is reduced, thereby improving monitoring efficiency and early warning success rate.

[0012] Preferably, the preset areas include at least grain production areas, grain storage areas, grain transportation areas, and grain sales areas, and the preset data include at least production data, processing data, transportation and logistics data, sales and market data, quality and safety data, and environmental and meteorological data.

[0013] Preferably, the historical data represents the relationship between the data indicator and the historical grain demand, and the historical data matching the historical grain demand with the predicted demand is used as the first screening criterion.

[0014] Preferably, the first variables include artificially controlled variables and uncontrollable variables, and the second variables are artificially controlled variables; taking the artificially controlled variables as monitoring objects facilitates adjusting each second variable to a corresponding data indicator.

[0015] Preferably, the changing trend of the second variable and the corresponding data indicator in the historical data is monitored. If the second variable changes against the trend, the corresponding food supply chain security early warning information is issued to the early warning center, and the early warning information includes the current value of the second variable and the corresponding data indicator; the adjustment method is to adjust the second variable to the corresponding data indicator. If the gap between the second variable and the corresponding data indicator widens, an early warning prompt is required.

[0016] Preferably, if the second variable changes against the trend, the changing trend of the second variable within the error interval continues to be monitored; if the value of the second variable at the end point of the error interval is closer to the corresponding data indicator in the historical data than the value at the starting point, the counter-trend change is regarded as a normal fluctuation of the second variable; if the value of the second variable at the starting point of the error interval is closer to the corresponding data indicator in the historical data than the value at the end point, the corresponding food supply chain security early warning information is issued to the early warning center; the data has a normal fluctuation range, and the allowable variation range of the second variable under normal circumstances is calibrated by the error interval and the values ​​of the second variable at the starting point and the end point of the error interval.

[0017] Preferably, if the second variable undergoes an adverse change, the changing trend of the second change within the error interval continues to be monitored; if the second variable undergoes a favorable change within the error interval, and the favorable change continues to the end point of the error interval, the adverse change is regarded as a normal fluctuation of the second variable; otherwise, the corresponding food supply chain security early warning information is issued to the early warning center; the allowable variation range of the second variable under normal circumstances is calibrated by the error interval and the changing trend of the second variable within the error interval, thereby further improving the fault tolerance of the second variable and reducing ineffective regulation of the second variable.

[0018] Preferably, if the starting point of the trend change of the second variable is between the starting point and the midpoint of the error interval, the counter-trend change is regarded as a normal fluctuation of the second variable; otherwise, the corresponding food supply chain security warning information is issued to the early warning center; the judgment of the starting point position of the trend change is added in the calibration process, which can increase the observation length of the second variable within the error interval, and help to obtain more accurate analysis results.

[0019] Preferably, the emergency plan includes an adjustment plan for the second variable to change to a corresponding data indicator in a natural manner; the second variable is directly regulated by the emergency plan to reduce the difficulty of data regulation.

[0020] Preferably, several segmented intervals are set, and the second variable gradually changes to the corresponding data indicator through the segmented intervals; for the second variable that deviates greatly from the data indicator, the second variable is gradually changed to the corresponding data indicator through the segmented intervals, so that the overall change trend of the second variable is controllable.

[0021] It can be seen from the above technical solution that the present invention has the following beneficial effects:

[0022] 1. The present invention provides a food supply chain security monitoring and early warning system based on big data analysis. Through the data acquisition module, the system can establish a complete and real-time updated food supply chain database; through the data analysis module, the changing trend of the food supply chain can be understood and future demand can be predicted; through the data monitoring module, the system can establish a benchmark model or benchmark data set, which is helpful to determine the most relevant and critical variables for monitoring and prediction, save monitoring procedures and improve early warning efficiency; through the early warning release module, potential safety risks can be timely understood; through the emergency response module, corresponding emergency plans are generated according to the early warning information, which enables the system to respond quickly to potential risks and threats, thereby reducing potential losses and impacts.

[0023] 2. If the second variable changes against the trend, that is, the second variable deviates from the corresponding data indicators in the historical data, the system will trigger the early warning mechanism. The adverse change can help the early warning center respond in time to prevent the problems in the food supply chain from further expanding; among them, the early warning information includes the current value of the second variable and the corresponding data indicators, which can help the early warning center better evaluate and respond to potential safety issues in the food supply chain.

[0024] 3. Furthermore, if the second variable changes against the trend, the system will continue to monitor the changing trend of the second variable within the error range to distinguish normal fluctuations from potential problems and improve accuracy; the system compares the values ​​corresponding to the starting and ending points of the second variable at the error range with the corresponding data indicators in the historical data. If the value corresponding to the end point is closer to the data indicator in the historical data than the value corresponding to the starting point, the system will regard the adverse change as a normal fluctuation of the second variable, thereby better protecting the security and stability of the food supply chain.

[0025] 4. Furthermore, if the second variable undergoes a counter-trend change, the system will observe the dynamic changes of the second variable. If the second variable undergoes a favorable change within the error interval and continues to the end point of the error interval, the system will regard the counter-trend change as a normal fluctuation of the second variable. By carefully observing and analyzing the changes in the second variable, the system can more accurately determine whether the counter-trend change is a normal fluctuation, thereby providing reliable early warning information.

[0026] 5. Based on the trend change, if the starting point of the trend change is between the starting point and the midpoint of the error interval, the system will regard the counter-trend change as a normal fluctuation of the second variable. By further subdividing and analyzing the starting point position of the trend change, the system can provide more accurate judgment and early warning.

[0027] 6. Through reasonable segmentation settings and gradual adjustments, the system can better control and manage the change process of the second variable to deal with problems in the supply chain and ensure the safety and stability of the food supply chain. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a structural block diagram of the food supply chain security monitoring and early warning system based on big data analysis according to the present invention;

[0029] Figure 2 This is a flow chart of the trend warning method described in the present invention;

[0030] Figure 3 This is a flow chart of a method for distinguishing normal data fluctuations according to the present invention;

[0031] Figure 4 This is a flow chart of another method for distinguishing normal data fluctuations according to the present invention;

[0032] Figure 5 This is a flow chart of a method for distinguishing normal fluctuations of data based on trend changes according to the present invention;

[0033] Figure 6 This is a flow chart of the method for distinguishing the starting point of the trend change according to the present invention;

[0034] Figure 7This is a flow chart of the threshold warning method described in the present invention;

[0035] Figure 8 This is a flow chart of the working method of the food supply chain security monitoring and early warning system based on big data analysis described in the present invention. DETAILED DESCRIPTION

[0036] The present invention is further explained below in conjunction with the accompanying drawings and specific embodiments.

[0037] Embodiment 1

[0038] like Figure 1 As shown, this embodiment provides a food supply chain security monitoring and early warning system based on big data analysis, including a data collection module 1, a data analysis module 2, a data monitoring module 3, an early warning release module 4 and an emergency response module 5.

[0039] Specifically, in this embodiment, the system includes a data collection module 1, whose function is to capture the preset data of each link of the food supply chain in a preset area, wherein the preset area at least includes food production areas, food storage areas, food transportation areas, and food sales areas. In general, the coverage of the preset area can be set according to actual needs, such as regions, provinces, cities, counties, etc., and the above-mentioned food production areas, food storage areas, etc. are all located in the preset area; the preset data at least includes production data, processing data, transportation and logistics data, sales and market data, quality and safety data, and environmental and meteorological data;

[0040] The production data refers to data related to planting, breeding and agricultural production, such as planting area, planting varieties, fertilization conditions, pesticide usage, crop yield, etc.; the processing data refers to data in the processing and handling of grain, such as production conditions of grain processing plants, processing parameters, product quality indicators, etc.; the transportation and logistics data refers to data in the transportation, storage and warehousing of grain, such as logistics routes, operation conditions of transportation tools, utilization rate of storage facilities, transportation time, etc.; the sales and market data refers to data on grain sales and market demand, such as market prices, sales volumes, market trends, consumer demand, etc.; the quality and safety data refers to data related to grain quality and safety, such as agricultural product quality testing data, pesticide residue testing data, food safety monitoring data, etc.; the environmental and meteorological data refers to data on the impact of environmental and meteorological factors on grain production, such as temperature, precipitation, soil moisture, meteorological disaster data, etc. It can be understood that the preset data is dynamic data and is updated at a preset frequency, wherein the preset frequency is set according to actual needs, such as seasonality and day and night. The sources of the above data include but are not limited to sensor technology, Internet of Things devices, farmer surveys, market research, and government statistics. For example, by installing sensor equipment in the production, processing, transportation, and storage of the grain supply chain, data such as temperature, humidity, and air pressure can be collected; by installing RFID tags on goods in the grain supply chain, automatic identification and tracking of goods can be achieved, and data from the transportation and storage of goods can be collected; through Internet technologies such as grain supply chain management systems and e-commerce platforms, transaction, inventory, and sales data can be collected.

[0041] Afterwards, the data acquisition module 1 integrates the preset data into the database. It can be understood that in the field of food supply chain, data platforms or databases are used to integrate and manage data, such as: relational databases (RDBMS): MySQL, Oracle, Microsoft SQL Server, etc., relational databases can provide structured data storage and query functions, which are convenient for storing and managing various data in the food supply chain; NoSQL databases: MongoDB, Cassandra, Redis, etc., NoSQL databases can store unstructured or semi-structured data and have good scalability and flexibility; Data warehouses: Amazon Redshift, Google BigQuery, Snowflake, etc., data warehouses can store large-scale data sets and support high-performance data query and analysis; Data Lakes: Apache Hadoop, Amazon S3, Azure Data Lake, etc., data lakes can store structured, semi-structured and unstructured raw data, which is convenient for subsequent data analysis and mining.

[0042] In addition to the above-mentioned databases and data platforms, there are also other solutions and cloud service providers. You can choose a suitable data platform or database based on specific needs, data types, and technical requirements.

[0043] The system also includes a data analysis module 2, whose function is to process and analyze preset data based on historical data, wherein:

[0044] The historical data includes the same type of preset data at past times, for example, production data, processing data, transportation and logistics data, sales and market data, quality and safety data, environment and meteorological data, etc. at past times.

[0045] The system uses big data analysis technology to clean the captured preset data and historical data, including processing missing values, outliers and duplicate values, which helps to ensure the quality and accuracy of the data for subsequent data analysis, modeling and decision-making. It is understandable that various data analysis tools and programming languages, such as Python's pandas library, R language, etc., can be used to support data cleaning operations.

[0046] Furthermore, food supply chain data may contain missing values, such as missing production records, missing transportation data, etc. Processing missing values ​​can avoid bias in the data during subsequent analysis. By filling missing values ​​or deleting data records containing missing values, the integrity of the data set can be ensured. In food supply chain data, outliers may be caused by measurement errors, abnormal situations, or other factors. Processing outliers can ensure the accuracy and consistency of the data. In food supply chain data, duplicate purchase records, duplicate transaction records, etc. will generate duplicate values. Processing duplicate values ​​can avoid repeated calculations of data or misleading during analysis.

[0047] Accordingly, the historical data is processed and analyzed, including observing the trend, seasonal changes, cyclical changes, etc. of the historical data. Specifically, the historical data is time series data, ensuring that the historical data includes a sufficient time span and frequency, and selecting an AR model according to the nature of the historical data. For example, by drawing a line graph of the time series data, the overall trend, seasonality, and cyclical characteristics of the historical data are observed, and an AR model with a trend term or a seasonal term is used. The AR model is an autoregressive model based on past observations, which assumes that future observations are related to past observations. The parameter values ​​in the AR model are estimated by maximum likelihood estimation or least squares method. The parameter values ​​reflect the autocorrelation relationship in the historical data. The AR model is fitted to the historical data of the historical grain quantity using the parameter values. Starting from the starting point of the historical data, predictions are made forward in sequence, using a recursive method, that is, each time an estimated value of a known observation is used to predict the next observation, to obtain future demand trends. The AR autoregressive model can predict food demand in different time periods, such as annual or quarterly food demand, and adjust and optimize according to actual conditions. The trained model is used to predict future food demand. Relevant factors include population growth rate, economic development trend, etc. The model gives corresponding prediction results. The prediction result here is the predicted demand.

[0048] The system also includes a data monitoring module 3, whose function is to monitor the changes of preset data, wherein the reference object is historical data.

[0049] The historical data can represent the relationship between the preset data at the past moment and the historical food demand. The preset data at the past moment is defined as a data indicator. The data indicator includes production data, processing data, transportation and logistics data, sales and market data, quality and safety data, environmental and meteorological data, etc. The preset data at the current moment needs to be compared with the data indicators corresponding to the historical food demand to regulate the preset data. The historical food demand is the actual food demand at the past moment. After determining the predicted demand, the historical data matching the preset demand is screened. The matching means that the preset demand is equal to or close to the historical food demand. The close means that the difference between the two is within the preset difference. If there are multiple historical data If there are any overlapped preset data or data indicators, the overlapping variables include production data, processing data, transportation and logistics data, sales and market data, quality and safety data, environment and meteorological data, one or more of the following: the overlapping variables are set as the first variable, and the remaining variables are set as the second variable. The remaining variables are specifically the remaining human-controlled variables. The second variable is the object of monitoring and adjustment of the system, which is manifested as adjusting the second variable at the current moment based on the data indicators in the historical data.

[0050] The so-called human control means that in modern agricultural management technology, most data can be adjusted manually. For example, meteorological data can be adjusted through a reasonable drainage system to provide a suitable growth environment. Reasonable farming operations and pest and disease control can also enhance the adaptability of crops to meteorological changes. Greenhouses or sheds can partially isolate the direct connection between the crop growth environment and external meteorological conditions, reduce the impact of meteorological changes on crops, and take measures such as sunshade nets and heat preservation sheds to adjust light and temperature. The above are all within the scope of human control.

[0051] For example, assuming that the predicted demand is the purchase volume of grain, the data monitoring module 3 will select data indicators equivalent to the purchase volume from the historical data, which include the supplier's supply, market price, seasonal factors, etc. Among them, assuming that the variables with the most overlap are the supplier's supply and seasonal factors, they will be set as the first variable, and the remaining market prices will be set as the second variable. The data monitoring module 3 monitors the value of the second variable and compares it with the market price in the historical data. The system can set early warning indicators to detect whether the market price has deviated.

[0052] The system also includes an early warning release module 4, whose function is to release corresponding food supply chain security early warning information to the early warning center.

[0053] like Figure 2As shown, this embodiment provides a trend warning method, which includes the following steps:

[0054] S101: monitoring the change trend of the second variable and the corresponding data indicator in the historical data;

[0055] S102: Determine whether the second variable undergoes an adverse change, where the adverse change refers to a change trend of the second variable deviating from a change trend of a corresponding data indicator;

[0056] S103: If yes, issue the corresponding food supply chain security warning information to the warning center; if no, continue to execute S101.

[0057] This method issues early warning by analyzing the changing trend of the second variable. It uses time series analysis or trend line fitting methods to observe the numerical changes of the second variable and sets corresponding trend early warning indicators. When the trend of the second variable deviates from the early warning indicator, an early warning is triggered.

[0058] It can be imagined that the changing trend includes trend changes and counter-trend changes. Trend changes and counter-trend changes require the introduction of data indicators corresponding to the second variable in historical data. Regardless of whether the value of the second variable increases or decreases, when the second variable is close to the corresponding data indicator, the system determines it as a trend change. When the second variable deviates from the corresponding data indicator, the system determines it as a counter-trend change. In the trend warning method, when the second variable undergoes a counter-trend change, an early warning is triggered.

[0059] In one case, the second variable is the amount of fertilizer applied. Take the amount of fertilizer applied to common crops as an example. For example, for corn, 20-30 kg of nitrogen fertilizer, 20-30 kg of phosphorus fertilizer and 20-30 kg of potassium fertilizer are applied per mu of land during the sowing period, and additional fertilizer is applied during the growth period according to the needs of the crop; for wheat, 20-30 kg of nitrogen fertilizer, 20-30 kg of phosphorus fertilizer and 20-30 kg of potassium fertilizer are applied before sowing, and additional fertilizer can be applied during the tillering and heading periods. The above data gives a reference. In actual operations, the current moment The amount of fertilizer applied is compared with the corresponding amount of fertilizer applied in the historical data. If the current amount of fertilizer applied deviates from the corresponding amount of fertilizer applied in the historical data, an early warning is triggered. For example, in the historical data, 20kg of nitrogen fertilizer, 20kg of phosphorus fertilizer, and 20kg of potassium fertilizer are applied per mu of land. In the preset data, 25kg of nitrogen fertilizer, 25kg of phosphorus fertilizer, and 25kg of potassium fertilizer are applied per mu of land. In this case, the amount of fertilizer applied needs to be reduced by 5kg each. In the subsequent adjustment process, if the amount of fertilizer applied increases, it will be regarded as a deviation from the historical data and an early warning will be triggered.

[0060] The early warning release module 4 will collect data related to the abnormal situation, such as the value of the second variable, the monitoring time, the degree of abnormality, etc. The generated early warning information includes but is not limited to the identification of the abnormal second variable, the time and place of the abnormality, and the evaluation of the degree of abnormality. The early warning release module 4 will transmit the generated early warning information to the early warning center in a timely manner, and convey the early warning information to relevant staff through appropriate communication channels. The early warning center can take corresponding actions based on the received early warning information, such as adjusting the fertilization plan, adjusting the procurement plan, coordinating with suppliers, or re-evaluating risks, etc.

[0061] The system also includes an emergency response module, whose function is to generate corresponding emergency plans.

[0062] When the early warning module 4 releases the food supply chain security early warning information to the early warning center, the early warning center will receive and analyze the early warning information. If the early warning center or relevant staff confirms that there is a potential supply chain security problem, the operation of the emergency response module 5 will be triggered. The emergency response module 5 will automatically generate a corresponding emergency plan based on the received early warning information. This process can be based on predefined rules and strategies, or it can use machine learning algorithms and decision models to generate emergency plans. The generated emergency plans include but are not limited to descriptions of emergency measures, division of responsibilities, resource allocation plans, coordination and communication strategies, and expected timetables and goals.

[0063] Furthermore, the emergency plan includes an adjustment plan for the second variable to change to the corresponding data indicator. For example, for corn, the current fertilizer amount is 20kg nitrogen fertilizer, 20kg phosphorus fertilizer and 20kg potassium fertilizer, and the corresponding fertilizer amount in the historical data is 30kg nitrogen fertilizer, 30kg phosphorus fertilizer and 30kg potassium fertilizer. The emergency plan is to increase 10kg nitrogen fertilizer, 10kg phosphorus fertilizer and 10kg potassium fertilizer.

[0064] Furthermore, the emergency plan includes an adjustment plan for the second variable to change to the corresponding data indicator in a natural way, and the adjustment plan also includes: setting a number of segmented intervals, and the second variable gradually changes to the corresponding data indicator through the segmented intervals.

[0065] Specifically, several segmented intervals are used to gradually adjust the second variable in line with the trend. The emergency plan module will gradually adjust the value of the second variable in each segmented interval according to the second variable and the corresponding data indicator in the warning information, so that it gradually approaches the corresponding data indicator.

[0066] For example, fractional fertilization is to apply the required nitrogen fertilizer, phosphorus fertilizer and potassium fertilizer into the soil in multiple times. For example, add segmented intervals in the above emergency plan. The first segment: add 5kg nitrogen fertilizer, 5kg phosphorus fertilizer and 5kg potassium fertilizer. The second segment: add 5kg nitrogen fertilizer, 5kg phosphorus fertilizer and 5kg potassium fertilizer. Through two segmented intervals, the 20kg nitrogen fertilizer, 20kg phosphorus fertilizer and 20kg potassium fertilizer are increased to 30kg nitrogen fertilizer, 30kg phosphorus fertilizer and 30kg potassium fertilizer.

[0067] Embodiment 2

[0068] like Figure 3 As shown, this embodiment is basically the same as the first embodiment, except that the second variable undergoes a reverse trend change further comprising:

[0069] If the second variable changes against the trend, then:

[0070] S201: Continue to monitor the change trend of the second variable within the error range;

[0071] S202: Determine whether the value of the second variable at the end point of the error interval is closer to the corresponding data indicator in the historical data than the value at the starting point;

[0072] S203: If yes, the counter-trend change is regarded as a normal fluctuation of the second variable;

[0073] S204: If not, the corresponding food supply chain security warning information is released to the early warning center.

[0074] Specifically, the normal fluctuation range of the data indicates the allowable variation range of the data within a certain period of time or under specific conditions. The error interval is [a, b], where a is the current moment, and a and b include a certain period of time. Starting from moment a, the variation trend of the second variable within the error interval [a, b] is detected. In Example 1, if the second variable changes against the trend within the error interval [a, b], an early warning is triggered. In this embodiment, the value x of the second variable at moment a and the value y of the second variable at moment b are first recorded. In the historical data, the data index at moment a is x', and the data index at moment b is y'. If |y-y'|>|x-x'|, an early warning is triggered.

[0075] Embodiment 3

[0076] like Figure 4 As shown, this embodiment is basically the same as the first embodiment, except that the second variable undergoes a reverse trend change further comprising:

[0077] If the second variable changes against the trend, then

[0078] S301: Continue to monitor the change trend of the second change within the error interval,

[0079] S302: Determine whether the second variable continues to change in a trend and the trend change continues to the end point of the error interval;

[0080] S303: If yes, the counter-trend change is regarded as a normal fluctuation of the second variable;

[0081] S304: If not, the corresponding food supply chain security warning information is released to the early warning center.

[0082] Specifically, the error interval is [a, b], a is the current moment, a and b include a certain time, and there is a moment c between moment a and moment b; starting from moment a, detect the change trend of the second variable in the error interval [a, b]. In Example 1, if the second variable changes against the trend in the error interval [a, b], an early warning is triggered. In this embodiment, if the second variable changes against the trend in the interval [a, c] and changes with the trend in the interval [c, b], the change against the trend is regarded as a normal fluctuation of the second variable; the numerical expression is: the value of the second variable at moment a is x, the value of the second variable at moment c is z, and the value of the second variable at moment b is y. In the historical data, the data index at moment a is x', the data index at moment c is z', and the data index at moment b is y'. If |z-z'|>|x-x'| and |z-z'|>|y-y'|, the change against the trend is regarded as a normal fluctuation of the second variable.

[0083] By further observing whether the second variable undergoes a trend change and continues to the end point of the error interval, the system can screen out counter-trend changes and distinguish normal fluctuations from abnormal situations, thereby improving the accuracy and credibility of the early warning.

[0084] Embodiment 4

[0085] like Figure 5 As shown, this embodiment is basically the same as the third embodiment, except that the second variable changes in the error interval in a trend manner and further includes:

[0086] S401: Determine whether the starting point where the second variable changes with the trend is between the starting point and the midpoint of the error interval;

[0087] S402: If yes, the counter-trend change is regarded as a normal fluctuation of the second variable;

[0088] S403: If not, the corresponding food supply chain security warning information is released to the early warning center.

[0089] Specifically, the error range is [a, b], where a is the current moment, there is a certain period of time between a and b, and there is a moment c between moment a and moment b; starting from moment a, the change trend of the second variable within the error range [a, b] is detected. In this embodiment, if the second variable shows a contrarian change in the interval [a, c] and a trend-following change in the interval [c, b], and a < c < (a + b) / 2, then the contrarian change is regarded as the normal fluctuation of the second variable.

[0090] This can further improve the accuracy of early warning, ensuring that only truly abnormal situations that require attention will trigger an early warning. This helps to take corresponding measures in a timely manner to address potential risks and problems, so as to ensure the safety and stability of the food supply chain.

[0091] Embodiment Five

[0092] This embodiment is basically the same as Embodiment One, except that this embodiment provides a threshold early warning method, which monitors the second variable based on a set threshold.

[0093] Specifically, the threshold is set based on the corresponding data indicators in historical data. In one case, the second variable is the food inventory. If the food inventory is lower than a certain threshold, an early warning of insufficient inventory is triggered.

[0094] Correspondingly, the emergency plan is to increase the inventory from the current moment to at least the threshold.

[0095] Furthermore, a segmented interval is added to the above emergency plan. The first segment: transfer a quantity of food inventory from Warehouse A; the second segment: transfer a quantity of food inventory from Warehouse B.

[0096] Specifically, as Figure 6 shown, the early warning method provided in this embodiment is as follows:

[0097] S501: Set the threshold and obtain the food inventory;

[0098] S502: Determine whether the food inventory is lower than the threshold;

[0099] S503: If so, send a corresponding food supply chain security early warning message to the early warning center. If not, continue to execute S301.

[0100] Embodiment Six

[0101] This embodiment is basically the same as the first embodiment, except that the artificially controlled variable in the first variable is monitored at the same time. It can be understood that the first variable is an overlapping variable in the preset data at the current moment and the historical data, that is, the first variable meets the preset demand at the current moment, but in actual applications, the first variable may also change. The artificially controlled variable in the first variable is extracted and set as the third variable, and the third variable is monitored.

[0102] like Figure 7 As shown, the early warning method provided in this embodiment is:

[0103] S601: extract the artificially controlled variable in the first variable and set it as the third variable, monitor the third variable, and set early warning indicators based on the third variable and historical data; it can be understood that the preset data is dynamic data, both the first variable and the second variable are likely to change, and there is also a possibility of deviation from the data indicator after the first variable changes, so the artificially controlled variable in the first variable will be monitored.

[0104] S602: Based on the monitoring situation and early warning indicators, issue corresponding food supply chain security early warning information to the early warning center;

[0105] S603: Generate a corresponding emergency plan based on the warning information.

[0106] Embodiment 7

[0107] like Figure 8 As shown, this embodiment is a working method corresponding to a food supply chain security monitoring and early warning system based on big data analysis according to the first embodiment, and includes the following working steps:

[0108] S701: Capturing preset data of each link of the food supply chain in a preset area, and integrating the preset data into a database;

[0109] S702: Processing and analyzing the preset data, and performing trend analysis on historical data to obtain a forecast of future food demand;

[0110] S703: Filter historical data that matches the predicted demand, and filter historical data with the most overlapping variables, set the overlapping variables in the preset data as first variables, set the remaining variables as second variables, monitor the second variables, and set early warning indicators according to the second variables and historical data;

[0111] S704: Release corresponding food supply chain security warning information to the early warning center based on monitoring conditions and early warning indicators;

[0112] S705: Generate a corresponding emergency plan based on the warning information.

[0113] The above description is only a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements without departing from the principle of the present invention, and these improvements should also be regarded as within the protection scope of the present invention.

Claims

1. A food supply chain security monitoring and early warning system based on big data analysis, characterized by: include: A data collection module is configured to capture preset data of each link of the food supply chain in a preset area and integrate the preset data into a database, wherein the preset data is dynamic data, the preset data includes a number of variables, and the preset data is updated at a preset frequency; A data analysis module is configured to process and analyze the preset data, and obtain a predicted demand for future food by performing trend analysis on historical data; The data monitoring module is configured to screen the historical data that matches the predicted demand, and screen the historical data that overlaps most with the preset data variables, set the variables in the preset data that overlap with the historical data as the first variable, set the remaining variables as the second variable, monitor the second variable, and set the early warning indicator according to the second variable and the historical data; extract the artificial control variable in the first variable and set it as the third variable, monitor the third variable, and set the early warning indicator according to the third variable and the historical data; The early warning release module is configured to release corresponding food supply chain security early warning information to the early warning center according to the monitoring situation and the early warning indicators, including: monitoring the change trend of the second variable and the corresponding data indicators in the historical data, and if the second variable changes against the trend, then releasing the corresponding food supply chain security early warning information to the early warning center, wherein the early warning information includes the current value of the second variable and the corresponding data indicators; The second variable undergoing an adverse change includes: if the second variable undergoes an adverse change, continuing to monitor the change trend of the second variable within the error interval, if the value of the second variable at the end point of the error interval is closer to the corresponding data indicator in the historical data than the value at the starting point, then the adverse change is regarded as a normal fluctuation of the second variable, if the value of the second variable at the starting point of the error interval is closer to the corresponding data indicator in the historical data than the value at the end point, then issuing corresponding food supply chain security warning information to the early warning center; or, the second variable undergoing an adverse change includes: if the second variable undergoes an adverse change, continuing to monitor the change trend of the second variable within the error interval, if the second variable undergoes a trend change within the error interval, and the trend change continues to the end point of the error interval, then the adverse change is regarded as a normal fluctuation of the second variable, otherwise, then issuing corresponding food supply chain security warning information to the early warning center; The emergency response module is configured to generate a corresponding emergency plan based on the warning information.

2. The food supply chain security monitoring and early warning system based on big data analysis according to claim 1 is characterized by: The preset areas include at least grain production areas, grain storage areas, grain transportation areas, and grain sales areas, and the preset data include at least production data, processing data, transportation and logistics data, sales and market data, quality and safety data, and environmental and meteorological data.

3. The food supply chain security monitoring and early warning system based on big data analysis according to claim 1 is characterized by: Screening historical data that matches preset needs also includes: the historical data includes several data indicators, the historical data represents the relationship between the data indicators and the historical food demand, and the historical data that matches the historical food demand and the predicted demand is used as the first screening criterion.

4. The food supply chain security monitoring and early warning system based on big data analysis according to claim 1 is characterized by: The first variable includes an artificially controlled variable and an uncontrollable variable, and the second variable is an artificially controlled variable.

5. The food supply chain security monitoring and early warning system based on big data analysis according to claim 1 is characterized by: The second variable undergoes a favorable change within the error interval and also includes: if the starting point of the favorable change of the second variable is between the starting point and the midpoint of the error interval, the counter-trend change is regarded as a normal fluctuation of the second variable; otherwise, corresponding food supply chain security warning information is issued to the warning center.

6. The food supply chain security monitoring and early warning system based on big data analysis according to claim 1 is characterized by: The corresponding emergency plans generated according to the early warning information also include: The emergency plan includes an adjustment plan for the second variable to change to the corresponding data indicator.

7. The food supply chain security monitoring and early warning system based on big data analysis according to claim 6 is characterized by: The adjustment plan also includes: A number of segmented intervals are set, and the second variable gradually changes to the corresponding data indicator through the segmented intervals.

Images