Agricultural product quality safety monitoring traceability tracking method

Abstract

The invention discloses an agricultural product quality safety monitoring traceability tracking method, which relates to the technical field of product quality safety monitoring and comprises the following steps of: constructing an agricultural product circulation relation chain, determining a sampling inspection proportion based on agricultural product categories and production factors, and performing differential multi-index sampling inspection on agricultural products input on farmers in the agricultural product circulation relation chain; predicting the residual safe storage time of the agricultural products based on a safe storage time prediction model, and performing marking early warning on high-risk batches; receiving an agricultural product ordering demand, further predicting an agricultural product deterioration probability, and judging whether to trigger to provide an optimized ordering strategy for a to-be-ordered user based on the agricultural product deterioration probability; and receiving complaint data of the ordering user, determining problem investigation and improvement requirements according to tracing query of the complaint data, and adjusting a spot check strategy. Full-chain differential monitoring and risk early warning from production to consumption are realized, and the quality management and control efficiency and the customer satisfaction are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of product quality and safety monitoring, and particularly to a method for monitoring, tracing and tracking the quality and safety of agricultural products. Background Technique

[0002] With the improvement of living standards, consumers are paying more and more attention to the quality and safety of agricultural products. Consumers hope to understand information such as the production process and quality status of the agricultural products they purchase in order to make more reassuring consumption decisions.

[0003] However, the entry and tracking of existing traceable data mainly rely on the self-discipline of market entities, and the quality is difficult to guarantee. At the same time, there are many standards for the traceability of agricultural product quality and safety at present, but there is a lack of effective communication and coordination between departments, resulting in problems such as overlapping content and inconsistent standards, leading to system incompatibility, duplicate construction and waste of resources; the traceability of agricultural product quality and safety mostly relies on post-event supervision and sampling inspection, lacking dynamic monitoring and risk prediction of the whole process of agricultural products.

[0004] Therefore, in view of the above problems, there is an urgent need for a method for monitoring, tracing and tracking the quality and safety of agricultural products. Summary of the Invention

[0005] In view of the deficiencies of the prior art, the present invention provides a method for monitoring, tracing and tracking the quality and safety of agricultural products, which solves the problems of difficult traceability, high loss and low user trust in the quality and safety of agricultural products.

[0006] To achieve the above object, the present invention is realized through the following technical solutions: A method for monitoring, tracing and tracking the quality and safety of agricultural products, including the following steps: Step S1, constructing an agricultural product circulation relationship chain, determining the sampling ratio based on the agricultural product category and production factors, and then conducting differentiated multi-index sampling inspections on the agricultural products entered by farmers in the agricultural product circulation relationship chain; Step S2, for the agricultural products that pass the differentiated multi-index sampling inspection, predicting the remaining safe storage time of the agricultural products based on the safe storage time prediction model, and then marking and warning high-risk batches; Step S3, receiving the agricultural product order demand, then identifying the user's receiving address and the agricultural product transportation end information, combining the user's receiving address, the agricultural product transportation end information and the predicted remaining safe storage time of the agricultural product to predict the probability of agricultural product spoilage, and feeding back the probability of agricultural product spoilage to the user waiting to place an order; Step S4, judging whether to trigger an optimized order placement strategy for the user waiting to place an order based on the probability of agricultural product spoilage, and the optimized order placement strategy includes receiving address optimization, transportation end optimization and agricultural product optimization; Step S5, receiving the complaint data of the order-placing user, and then determining the problem investigation and improvement requirements according to the traceability query of the complaint data, and adjusting the sampling inspection strategy.

[0007] Furthermore, each node in the agricultural product circulation relationship chain includes farmers, suppliers, sales points, transportation ends and order-placing users.

[0008] Further, step S1 is specifically analyzed as follows: identify the agricultural product categories entered by farmers, then identify the category risk coefficients based on the agricultural product categories, identify the production risk coefficients based on production factors, and then determine the sampling ratio using the category risk coefficients and production risk coefficients; determine multi-dimensional sampling indicators according to the characteristics of the agricultural product categories, and then randomly select corresponding numbers of samples from the batches of agricultural products entered by farmers according to the determined sampling ratio, and judge whether the agricultural products can enter the subsequent circulation link according to the test results of the selected samples.

[0009] Further, step S2 is specifically analyzed as follows: obtain the agricultural product characteristic data based on the tests of the selected samples, where the agricultural product characteristic data includes basic attributes, physicochemical properties, and physiological properties; identify the transportation end data in the agricultural product circulation relationship chain, where the transportation end data includes transportation equipment information and transportation environment information; use the agricultural product characteristic data and the transportation end data as the input of the safe storage time prediction model, and output the remaining safe storage time of the agricultural products. Compare the remaining safe storage time of the agricultural products with the safe storage time threshold. When the predicted remaining safe storage time of the agricultural products is lower than the safe storage time threshold, mark this batch of agricultural products as a high-risk batch.

[0010] Further, step S3 is specifically analyzed as follows: use the order receiving port to receive the agricultural product purchase orders from the placing users in real time, and extract the delivery address information filled in by the users from the order details; identify the agricultural product transportation end information related to this order according to the constructed agricultural product circulation relationship. The transportation end information includes the name of the logistics enterprise responsible for this transportation, the transportation method, the storage environment, the expected transportation route and duration, and extract the geographical feature data in the expected transportation route; call the predicted remaining safe storage time of this batch of agricultural products, and then output the spoilage probability value of this batch of agricultural products during the transportation to the user's delivery address according to the user's delivery address, the agricultural product transportation end information, and the remaining safe storage time of the agricultural products.

[0011] Further, step S4 is specifically analyzed as follows: When the spoilage probability is greater than or equal to the spoilage probability threshold, it triggers to provide an optimized order placement strategy for the user to be ordered. The optimization of the delivery address is specifically as follows: Retrieve the historical order records of the user to be ordered, extract the historical delivery address information of the user to be ordered from the historical order records, output the spoilage probability values respectively based on the historical delivery address information of the user to be ordered, mark the historical delivery addresses with spoilage probabilities lower than the spoilage probability threshold as optimized recommended addresses, and recommend them to the user to be ordered; The optimization of the transportation end is specifically as follows: Retrieve the remaining transportation methods and storage environments uploaded by the transportation end according to the established agricultural product circulation relationship, and then adjust the corresponding estimated transportation routes and durations to form an optimized set of transportation end information. Output the spoilage probability values respectively based on the optimized set of transportation end information, mark the transportation end information with spoilage probabilities lower than the spoilage probability threshold as optimized recommended transportation end information, and recommend it to the user to be ordered; The optimization of the agricultural product is specifically as follows: Identify the agricultural product to be ordered by the user to be ordered, then retrieve the similar agricultural products of this agricultural product, generate the corresponding spoilage probability values based on the similar agricultural products, mark the similar agricultural products with spoilage probabilities lower than the spoilage probability threshold as optimized recommended agricultural products, and recommend them to the user to be ordered.

[0012] Further, step S5 is specifically analyzed as follows: When it is detected that there is complaint data of the user who placed the order, check the spoilage probability feedback when the user who placed the order ordered the agricultural product. When the spoilage probability is greater than or equal to the spoilage probability threshold, mark this user who placed the order as a suspected malicious order placer, and trigger the artificial customer service to soothe the emotions of the user who placed the order; When the spoilage probability is lower than the spoilage probability threshold, trigger the problem tracing process, retrieve the sales records of the agricultural products in the same batch, and check whether there is complaint data of other users. If there is complaint data of other users, respectively check whether there is complaint data of other categories of agricultural product orders at other nodes in the agricultural product circulation relationship chain. If it is detected that there is complaint data of other categories of agricultural product orders at a certain node, then mark the corresponding user at this node with a risk and increase the sampling ratio of this user; If there is no complaint data of other categories of agricultural product orders at other nodes in the agricultural product circulation relationship chain for this batch of agricultural products, then increase the sampling ratio of the corresponding category of this batch of agricultural products.

[0013] The present invention has the following beneficial effects:

[0014] The method for monitoring, tracing and tracking the quality and safety of agricultural products dynamically adjusts the sampling ratio according to the categories of agricultural products and production factors, avoiding the inefficient detection of "one-size-fits-all", ensuring that high-risk categories are more strictly monitored, and at the same time reducing the detection costs of compliant farmers; by constructing a relationship chain for the circulation of agricultural products, the whole process traceability from farmers to consumers is realized. Once quality problems occur, the problem links can be quickly located, the recall scope can be reduced, and losses can be minimized; based on the model, the remaining safe storage time of agricultural products is predicted, high-risk batches are marked in advance, and farmers or merchants are guided to promote or remove products from shelves in a timely manner, reducing losses caused by expiration; combined with transportation conditions and the user's delivery address, the deterioration risk of agricultural products is dynamically evaluated, and the information is fed back to consumers to avoid quality disputes caused by transportation delays or improper storage; consumers can view the traceability information and deterioration probability of agricultural products when placing orders, improving the transparency of purchase decisions, reducing the complaint rate, and enhancing the user experience; by tracing complaint data, the root causes of problems are quickly located, farmers are guided to improve production processes or logistics plans, and the sampling ratio and indicators are dynamically adjusted according to complaint data, realizing the precise allocation of regulatory resources and improving regulatory efficiency. Description of the Drawings

[0015] Figure 1 It is a flow chart of a method for monitoring, tracing and tracking the quality and safety of agricultural products according to the present invention. Detailed Embodiment

[0016] In the embodiment of the present application, through a method for monitoring, tracing and tracking the quality and safety of agricultural products, differential monitoring and risk warning of the whole chain from production to consumption are realized, significantly improving the quality control efficiency and consumer satisfaction.

[0017] The general idea of the embodiment of the present application is as follows: through user identity verification and the construction of a circulation relationship chain, the sources and flows of agricultural products are clarified, laying a foundation for the whole process traceability; differential sampling is carried out based on the categories of agricultural products and production factors to ensure that the quality of agricultural products meets the standards; a risk assessment is carried out on the agricultural products that pass the sampling inspection by using a safe storage time prediction model, and high-risk batches are marked; the deterioration probability is predicted by combining the user's delivery address and transportation information, providing a reference for consumers; according to the complaint data of consumers, the root causes of problems are traced, the sampling strategy is adjusted, and the whole system is continuously optimized to form a closed-loop quality and safety management system.

[0018] Please refer to Figure 1, an embodiment of the present invention provides a technical solution: a method for monitoring, tracing, and tracking the quality and safety of agricultural products, including the following steps: Step S1, construct a circulation relationship chain of agricultural products, determine the sampling ratio based on the categories of agricultural products and production factors, and then conduct differentiated multi-index sampling inspections on the agricultural products entered by farmers in the circulation relationship chain of agricultural products; Step S2, for the agricultural products that pass the differentiated multi-index sampling inspection, predict the remaining safe storage time of the agricultural products based on the safe storage time prediction model, and then mark and give early warnings to high-risk batches; Step S3, receive the order demand for agricultural products, then identify the user's delivery address and the information of the agricultural product transportation end, combine the user's delivery address, the information of the agricultural product transportation end, and the predicted remaining safe storage time of the agricultural products to predict the spoilage probability of the agricultural products, and feedback the spoilage probability of the agricultural products to the user waiting to place an order; Step S4, based on the spoilage probability of the agricultural products, determine whether to trigger the provision of an optimized order placement strategy for the user waiting to place an order. The optimized order placement strategy includes delivery address optimization, transportation end optimization, and agricultural product optimization; Step S5, receive the complaint data of the order-placing user, then determine the problem investigation and improvement requirements according to the traceability query of the complaint data, and adjust the sampling inspection strategy.

[0019] Specifically, each node in the circulation relationship chain of agricultural products includes farmers, suppliers, sales points, transportation ends, and order-placing users.

[0020] Step S1 is specifically analyzed as follows: Verify the certification application information submitted by the user based on the review strategy to determine the node to which the user belongs in the circulation relationship chain of agricultural products. The review strategy includes farmer identity verification strategy, supplier identity verification strategy, sales point identity verification strategy, transportation end identity verification strategy, and order-placing user identity verification strategy; Obtain the agricultural product circulation-related information entered by each node in the circulation relationship chain of agricultural products, and then establish the relationship chain between each node according to the agricultural product circulation-related information entered by each node. The specific relationship chain is a complete agricultural product circulation relationship map, monitor the circulation of agricultural products and the occurrence of new transactions in real time, and update the information in the relationship chain.

[0021] Identify the categories of agricultural products entered by farmers, then identify the category risk coefficient based on the categories of agricultural products, identify the production risk coefficient based on production factors, and then determine the sampling ratio using the category risk coefficient and the production risk coefficient; Determine multi-dimensional sampling inspection indicators according to the characteristics of the categories of agricultural products, and then randomly select corresponding numbers of samples from the batches of agricultural products entered by farmers according to the determined sampling ratio, and judge whether the agricultural products can enter the subsequent circulation link according to the test results of the selected samples.

[0022] In this implementation plan, the farmer identity verification strategy specifically includes: personal information verification, which requires farmers to provide valid identity documents such as ID cards and household registers, and verifies the basic information of their identities, such as name, ID number, date of birth, etc., by comparing with the population information database of the public security department; verification of agricultural production qualifications, which identifies whether farmers have relevant agricultural production qualification certificates, such as land contract contracts and membership certificates of agricultural cooperatives. For land contract contracts, verify with the local land management department or the village committee to confirm the legality and authenticity of their land operations; on-site inspection and verification, where staff conduct on-site inspections at the agricultural production sites of farmers to verify their actual planting or breeding situations, and compare the types and areas of planted crops, and the quantities and breeds of farmed livestock and poultry with the information declared by farmers to determine whether it is true; verification of historical transaction records, by querying the transaction records on the agricultural product trading platform, identifying information such as the types and quantities of traded agricultural products and trading partners, and determining whether it conforms to the production and operation characteristics of farmers. It is also possible to trace the upstream and downstream transactions of farmers to verify the authenticity of the transactions and the identity information of farmers.

[0023] The supplier identity verification strategy specifically includes: enterprise qualification review, verifying the authenticity and validity of the business license provided by the supplier through the enterprise information query system of the administrative department for industry and commerce; verification of the business premises, where staff conduct on-site inspections at the business premises of the supplier to verify whether its warehouses, office facilities, etc. meet the conditions for supplying agricultural products; query of industry reputation, using the information in the industry credit database or a third-party credit assessment agency to obtain the credit records and industry reputation of the supplier, and understanding whether it has any bad records such as defaults or quality problems.

[0024] The sales point identity verification strategy specifically includes: review of store qualifications, verifying the authenticity and validity of the business license and relevant business licenses provided by the sales point; on-site inspection of the store, where staff conduct on-site inspections of the sales point to verify whether the store decoration, display, commodity sales, etc. conform to normal business characteristics; query of consumer evaluations and complaints, obtaining the consumer evaluations and complaints of the sales point through online evaluation platforms to understand its reputation among consumers and service quality.

[0025] The transportation end identity verification strategy specifically includes: review of transportation enterprise qualifications, verifying the road transport business license provided by the transportation enterprise, as well as the vehicle driving license and transportation license of the transport vehicle; inspection of transportation equipment, conducting on-site inspections of the transport vehicle, and at the same time querying the vehicle driving track and transportation records through the system to verify the authenticity and standardization of its transportation activities; review of driver qualifications, verifying the authenticity and validity of the driver's license and professional qualification certificate provided by the transportation enterprise, and retrieving traffic violation and accident records to evaluate its driving safety and reliability.

[0026] The identity verification policy for placing orders specifically includes: reviewing account registration information to verify the real-name authentication information of the order placer; examining historical transaction records, extracting the historical transaction records of the order placer, and identifying whether there are abnormal transaction behaviors, such as placing large orders frequently or concentrating on placing orders during specific time periods, to further verify their identity and transaction purpose; verifying the delivery address, verifying the authenticity and validity of the delivery address provided by the user.

[0027] The specific steps for identifying the category risk coefficient based on agricultural product categories are analyzed as follows: classifying agricultural products, such as classifying them into categories like fruits and vegetables, grains and oils, livestock and poultry meats, etc.; collecting data such as the frequency and harm degree of quality problems that various agricultural products have appeared in the past market circulation; combining expert experience and industry standards to evaluate the risk levels of different categories of agricultural products and determine the category risk coefficient. An example of the calculation formula for the category risk coefficient is: R c = k1*f1 + k2*f2 + k3*f3, where R in the formula c represents the category risk coefficient, k1, k2, and k3 respectively represent weight values, and f1, f2, and f3 respectively represent the frequency of past quality problems, harm degree, and market circulation volume.

[0028] The specific steps for identifying the production risk coefficient based on production factors are analyzed as follows: determining production factors, such as planting / breeding environment, use of inputs (pesticides, fertilizers, feeds, etc.), production technology level, and producer qualifications; conducting risk assessments on each factor, and then comprehensively obtaining the production risk coefficient. An example of the calculation formula for the production risk coefficient is: R p = w1*g1 + w2*g2 + w3*g3, where R in the formula p represents the production risk coefficient, w1, w2, and w3 respectively represent weight values, and g1, g2, and g3 respectively represent the compliance score of input use, production technology level, and suitability score of the planting / breeding environment.

[0029] The specific steps for determining the sampling inspection ratio using the category risk coefficient and the production risk coefficient are as follows: comprehensively considering the category risk coefficient and the production risk coefficient, and determining the sampling inspection ratio according to the magnitudes of the two values. When both the category risk coefficient and the production risk coefficient are relatively high, the sampling inspection ratio is correspondingly increased, and vice versa. An example of the specific calculation formula for the sampling inspection ratio is: χ = φ*(R c + R p ), where χ represents the sampling inspection ratio, φ represents the adjustment coefficient, which is determined by the actual production situation and has a value range between 0 and 1, and is used to adjust to ensure that the sampling inspection ratio is within a reasonable range.

[0030] The specific steps for determining multi-dimensional sampling inspection indicators according to the characteristics of agricultural product categories are as follows: The characteristics of agricultural product categories include physical characteristics (appearance, size, shape, color, etc.), chemical characteristics (nutritional components, pesticide residues, veterinary drug residues, heavy metal content, etc.), biological characteristics (microbial content, pest and disease conditions, etc.), and preservation characteristics (shelf life, storage conditions, etc.). Examples of specific multi-dimensional sampling inspection indicators are as follows: For fruits and vegetables, the multi-dimensional sampling inspection indicators are whether there are damages to the appearance, whether the pesticide residues exceed the standard, and whether the microbial content is qualified. For grain and oil products, the multi-dimensional sampling inspection indicators are the detection of nutritional components and the content of harmful substances such as aflatoxin. For livestock and poultry meats, the multi-dimensional sampling inspection indicators are veterinary drug residues, clenbuterol content, and freshness.

[0031] The detection implementation analysis is as follows: Send the extracted samples to a qualified professional testing institution, and detect each indicator according to the established testing standards and methods to obtain a detailed testing report issued by the testing institution. Record the testing results of each sample on each sampling inspection indicator. For example, for pesticide residue detection, clarify the names and residual amounts of various pesticides detected, and for microbial detection, report the detection situation and content of various microorganisms; Feed back the testing results to farmers and relevant regulatory departments. If the testing results are qualified, the agricultural products can enter the subsequent circulation link normally. If unqualified situations occur, further analyze the reasons for non-conformity. For example, excessive pesticide residues may be due to improper use of pesticides or failure to comply with the safety interval. Take corresponding measures for the problems, such as training and guiding farmers, punishing illegal acts, and tracing and handling the agricultural products of the same batch that have not been sampled to prevent unqualified agricultural products from entering the market.

[0032] Verify the user identity through strict audit policies to ensure the authenticity and reliability of information at each node. After constructing the relationship graph, the entire circulation process of agricultural products from the production source to the final consumers is clearly presented. Information on each link, such as the category, quantity, price, and circulation time of agricultural products, can be accurately traced, greatly enhancing information transparency and strengthening consumers' confidence in the quality and safety of agricultural products. Once quality problems occur in agricultural products, they can be quickly traced back to the source of the problem along the relationship chain. Whether it is the production link of farmers, such as improper use of pesticides, or poor storage conditions at the transportation end, it can be accurately located, facilitating timely measures to recall problem products, preventing the expansion of hazards, and at the same time clarifying the responsibility attribution and dealing with the responsible parties accordingly, effectively guaranteeing the quality and safety of agricultural products. For nodes such as suppliers and sales points, the clear relationship chain helps to reasonably arrange inventory, optimize logistics distribution, and accurately predict market demand. For example, suppliers can adjust the supply plan in a timely manner based on the sales data and inventory information of sales points, reducing inventory backlogs or shortages, lowering operating costs, and improving the efficiency and effectiveness of the entire supply chain. The audit policy prompts each node to standardize its own behavior and meet market access standards. Farmers need to provide true and valid production qualifications, and the transportation end must ensure that the transportation conditions meet the requirements, etc., which helps to eliminate unqualified practitioners, purify the market environment, and promote the healthy and orderly development of the agricultural product industry. Real-time monitoring and updating of relationship chain information enable each node to quickly perceive market dynamics, such as changes in consumer demand and adjustments in policies and regulations. Based on this information, enterprises can carry out business innovation, such as developing new agricultural product sales models and optimizing transportation plans, to better adapt to market changes and enhance market competitiveness.

[0033] Determine the sampling proportion by identifying the category risk coefficient and production risk coefficient, which can conduct differential sampling for agricultural products with different risk levels, avoiding the blindness of "one-size-fits-all" sampling and improving the utilization efficiency of sampling resources; determine multi-dimensional sampling indicators according to the characteristics of agricultural product categories, which can comprehensively and meticulously detect the quality of agricultural products, timely discover potential quality problems, and ensure the food safety of consumers.

[0034] Specifically, step S2 is specifically analyzed as follows: Obtain agricultural product characteristic data based on the detection of the extracted samples. The agricultural product characteristic data includes basic attributes, physicochemical properties, and physiological properties; identify the transportation end data in the agricultural product circulation relationship chain. The transportation end data includes transportation equipment information and transportation environment information; use the agricultural product characteristic data and transportation end data as the input of the safe storage time prediction model, and output the remaining safe storage time of the agricultural products. Compare the remaining safe storage time of the agricultural products with the safe storage time threshold. When the predicted remaining safe storage time of the agricultural products is lower than the safe storage time threshold, mark this batch of agricultural products as a high-risk batch.

[0035] In this implementation plan, the basic attributes include, but are not limited to, the variety, origin, and grade of agricultural products, which are entered by farmers or relevant nodes in the agricultural product circulation relationship chain and stored and queried through the system database. The specific variety can be quantified by using a classification coding method. For example, different varieties of apples are assigned different codes. The origin can be represented by geographical coordinates or area codes, and the grade can be numerically coded according to relevant standards, such as grade 1 is 1, grade 2 is 2, etc.

[0036] The physical and chemical properties include, but are not limited to, appearance (such as color, shape, size, etc.), nutrient components (contents of protein, fat, carbohydrates, etc.), pesticide residues, veterinary drug residues, and heavy metal contents. The appearance is obtained through visual inspection or image recognition technology, and the nutrient components, pesticide residues, veterinary drug residues, heavy metal contents, etc. are analyzed and detected by professional laboratory testing equipment. The color of the appearance can be quantified through a color space model, such as RGB values or Lab values. The shape and size can be represented by dimension measurement data, and the nutrient components, pesticide residues, veterinary drug residues, heavy metal contents, etc. are expressed as specific concentration or content values according to the test results.

[0037] The physiological properties include, but are not limited to, respiration intensity, ethylene release amount, and maturity. The respiration intensity and ethylene release amount are measured using specialized gas detection instruments. The maturity is determined through sensory evaluation, physical and chemical index detection (such as hardness, soluble solid content, etc.) or non-destructive testing techniques (such as near-infrared spectroscopy analysis). The respiration intensity is expressed as the amount of oxygen absorbed or carbon dioxide released by the agricultural product per unit mass per unit time, and the ethylene release amount is quantified as the volume or mass of ethylene released by the agricultural product per unit mass per unit time. The maturity can establish corresponding quantification indicators according to different detection methods. For example, hardness is expressed in Newtons, and the soluble solid content is expressed as a percentage.

[0038] The transportation equipment information includes, but is not limited to, the type of transportation vehicle, the model of refrigeration equipment, and the carriage volume. The relevant information is entered by the transportation side in the agricultural product circulation relationship chain, or automatically obtained and uploaded to the system through the electronic tags of transportation equipment, vehicle-mounted information systems, etc. The type of transportation vehicle can be represented by a code. For example, a truck is 1, a refrigerated truck is 2, etc. The model of the refrigeration equipment can directly adopt the model number of the equipment, and the carriage volume is quantified in cubic meters.

[0039] The transportation environment information includes, but is not limited to, transportation temperature, humidity, vibration conditions, and light intensity. Data is collected in real-time by installing temperature and humidity sensors, vibration sensors, light sensors, etc. in the transportation equipment and uploaded to the agricultural product circulation relationship chain system. The transportation temperature is in degrees Celsius, the humidity is expressed as a percentage, the vibration conditions can be quantified by parameters such as the acceleration or vibration frequency measured by the sensor, and the light intensity is in lux.

[0040] The steps for constructing the safety storage time prediction model are as follows: Collect a large amount of agricultural product characteristic data, transportation end data, and the corresponding actual safety storage time data as the sample set for model training; Screen out the characteristic parameters that have a significant impact on the safety storage time from the collected data and remove irrelevant or redundant features; According to the data characteristics and problem requirements, select an appropriate prediction model, such as regression models (linear regression, non-linear regression, etc.), neural network models, support vector machine models, etc.; Use the selected model, with the agricultural product characteristic data and transportation end data as inputs and the actual safety storage time as the output, to train the model and adjust the model parameters so that the model can accurately fit the relationship between the input and output; Use the validation data set to evaluate the trained model, calculate indicators such as the prediction accuracy rate and mean square error of the model, and evaluate the performance of the model; According to the evaluation results, optimize and adjust the model, such as adjusting the model parameters, adding or reducing features, etc., to improve the prediction accuracy and generalization ability of the model. An example of the expression of the safety storage time prediction model is:

[0041] In the formula, T represents the safety storage time, β0, β1, β2,..., β m , α1, α2,..., α h represent the model coefficients, m and h respectively represent the sample numbers of the agricultural product characteristic data and the transportation end data, represents the error term, η1, η2,..., η m , μ1, μ2,..., μ h respectively represent the agricultural product characteristic data and the transportation end data.

[0042] The method for setting the safety storage time threshold is as follows: Based on the safety storage period specified in the relevant industry standards or enterprise standards of agricultural products, make appropriate adjustments in combination with the actual situation to determine the safety storage time threshold; It is also possible to conduct experiments on agricultural products of different categories, different origins, and different grades under various common transportation and storage conditions, observe the quality change rules, and determine the longest storage time that can be achieved while ensuring the quality and safety of agricultural products, and use this as the safety storage time threshold; Or collect the actual storage time and quality status data of agricultural products during the previous circulation process, analyze the time distribution of quality problems of agricultural products in different situations, and determine a reasonable safety storage time threshold according to the statistical results. For example, the average value or median of the time when quality problems occur can be used as a reference for the threshold.

[0043] By comprehensively considering the characteristic data of agricultural products and the data at the transportation end to predict the safe storage time, it is possible to more comprehensively and accurately evaluate the quality status of agricultural products under current conditions and the remaining safe storage time, providing strong support for the quality monitoring of agricultural products; comparing the prediction results with the safe storage time threshold can timely detect batches of agricultural products that may have quality risks, facilitating the adoption of corresponding measures such as strengthening supervision and early handling, reducing losses caused by the deterioration of agricultural products and food safety risks; providing a scientific basis for the storage, transportation and other links in the circulation process of agricultural products, helping to optimize the management process and improve the efficiency and quality of the agricultural product supply chain.

[0044] Specifically, step S3 is specifically analyzed as follows: Use the order receiving port to receive the agricultural product purchase order from the placing user in real time, and extract the delivery address information filled in by the user from the order details; Based on the constructed agricultural product circulation relationship, identify the agricultural product transportation end information related to this order. The transportation end information includes the name of the logistics enterprise responsible for this transportation, the transportation method, the storage environment, the expected transportation route and duration, and extract the geographical feature data in the expected transportation route; Call the predicted remaining safe storage time of this batch of agricultural products, and then output the deterioration probability value of this batch of agricultural products during the transportation to the user's delivery address based on the user's delivery address, the agricultural product transportation end information and the remaining safe storage time of the agricultural products.

[0045] In this implementation plan, the geographical feature data includes but is not limited to landforms (such as plains, mountains, hills, etc.), climate zones (such as temperate zones, subtropical zones, tropical zones, etc.), altitude, road conditions (such as road condition grades, congestion conditions, etc.), and meteorological data such as humidity and temperature in the areas passed by the transportation route.

[0046] The steps for constructing the spoilage probability model are as follows: Collect a large amount of data related to agricultural product transportation, including data such as different transportation modes, storage environments, estimated transportation routes, durations, geographical feature data, the remaining safe storage time of agricultural products, and the corresponding spoilage situations of agricultural products; Select the features that have a significant impact on the spoilage probability from the collected data, such as the speed and stability of the transportation mode, the temperature and humidity of the storage environment, the length and road conditions of the estimated transportation route, the climate conditions in the geographical feature data, the remaining safe storage time of agricultural products, etc., and perform quantization processing; According to the data characteristics and the nature of the problem, select a suitable prediction model, such as a multiple linear regression model, a decision tree model, a neural network model, etc.; Use the sorted data to train the selected model. By adjusting the parameters of the model, make the model accurately fit the data and minimize the error between the prediction result and the actual spoilage situation. During the training process, adopt methods such as cross-validation to prevent the model from overfitting and improve the generalization ability of the model; Use an independent test data set to evaluate the trained model, calculate indicators such as the accuracy rate, recall rate, and mean square error of the model, and verify the effectiveness and reliability of the model. If the model evaluation result is not ideal, adjust and optimize the model until a satisfactory effect is achieved. An example of the expression for obtaining the spoilage probability is: ξ = γ0 + γ1 * λ1 + γ2 * λ2 +... + γ i *λ i + θ, where ξ represents the spoilage probability, γ0, γ1, γ2,..., γ i represent the model coefficients, θ represents the error term, and λ1, λ2,..., λ i represent the model input features respectively, and i represents the total number of input features.

[0047] By obtaining order information, transportation end information, and the remaining safe storage time of agricultural products in real time, the spoilage risk of agricultural products during transportation can be accurately evaluated, which helps to take measures in advance, such as accelerating the transportation speed, improving the storage conditions, etc., to ensure the quality of agricultural products delivered to users; Provide users with the spoilage probability value of agricultural products, enabling users to have a clearer understanding of the quality of the agricultural products they purchase, enhancing users' trust in the quality of agricultural products, and improving the user purchase experience; The comprehensive collection and analysis of transportation end information can help logistics enterprises optimize transportation routes and transportation modes, improve transportation efficiency, and reduce the loss of agricultural products during transportation.

[0048] Specifically, step S4 is specifically analyzed as follows: When the deterioration probability is greater than or equal to the deterioration probability threshold, trigger to provide an optimized order placement strategy for the user to be ordered. The optimization of the delivery address is specifically as follows: Retrieve the historical order records of the user to be ordered, extract the historical delivery address information of the user to be ordered from the historical order records, output the deterioration probability values respectively based on the historical delivery address information of the user to be ordered, mark the historical delivery addresses with deterioration probabilities lower than the deterioration probability threshold as optimized recommended addresses, and recommend them to the user to be ordered; The optimization of the transportation end is specifically as follows: Retrieve the remaining transportation methods and storage environments uploaded by the transportation end according to the constructed agricultural product circulation relationship, and then adjust the corresponding estimated transportation routes and durations, combine them to form an optimized set of transportation end information, output the deterioration probability values respectively based on the optimized set of transportation end information, mark the transportation end information with deterioration probabilities lower than the deterioration probability threshold as optimized recommended transportation end information, and recommend it to the user to be ordered; The optimization of the agricultural product is specifically as follows: Identify the agricultural product to be ordered by the user to be ordered, then retrieve the similar agricultural products of this agricultural product, generate the corresponding deterioration probability values based on the similar agricultural products, mark the similar agricultural products with deterioration probabilities lower than the deterioration probability threshold as optimized recommended agricultural products, and recommend them to the user to be ordered.

[0049] In this implementation plan, when the spoilage probability reaches the threshold, an optimized ordering strategy is provided for the users to be ordered from three aspects: the receiving address, the transportation end, and the agricultural products. By optimizing the three key factors of the receiving address, the transportation end, and the agricultural products, options with a spoilage probability lower than the threshold are selected and recommended to the users, which can effectively reduce the possibility of users receiving spoiled agricultural products, ensure the quality and safety of agricultural products, and reduce the losses suffered by users due to the spoilage of agricultural products; providing a targeted optimized ordering strategy for the users to be ordered, meeting the users' expectations for the quality of agricultural products, enhancing the users' trust and satisfaction with the platform, helping to improve the users' shopping experience, increasing the users' loyalty and the competitiveness of the platform; in terms of optimizing the transportation end, retrieving and adjusting the transportation mode, storage environment, estimated transportation route and duration, which can reasonably allocate logistics resources, improve transportation efficiency, reduce transportation costs, and at the same time reduce the loss of agricultural products during transportation, realizing more effective utilization of resources; in terms of optimizing agricultural products, by identifying similar agricultural products and recommending products with a low spoilage probability, it not only provides more choices for users, but also helps to promote other high-quality agricultural products, expand the sales scope, increase the sales volume of agricultural products, and has positive economic benefits for both agricultural product suppliers and the platform; based on the analysis and mining of data such as the users' historical order records and the circulation relationship of agricultural products, it is possible to deeply understand the users' needs and the circulation of agricultural products, provide valuable data support for the platform, so that the platform can make more scientific and reasonable decisions, and continuously optimize the service and operation strategies; when the spoilage probability reaches the threshold, the optimization strategy is triggered, which plays a role in risk warning and prevention and control, taking measures in advance to reduce risks, avoiding potential disputes and negative impacts caused by the spoilage problem of agricultural products, and maintaining the good image and reputation of the platform.

[0050] Specifically, the specific analysis of step S5 is as follows: when detecting the complaint data of the ordering user, check the spoilage probability feedback when the ordering user orders agricultural products. When the spoilage probability is greater than or equal to the spoilage probability threshold, mark the ordering user as a suspected malicious ordering user and trigger the artificial customer service to soothe the emotions of the ordering user; when the spoilage probability is lower than the spoilage probability threshold, trigger the problem tracing process, retrieve the sales records of the same batch of agricultural products, and check whether there is any complaint data of other users. If there is complaint data of other users, respectively detect whether there is any complaint data of other categories of agricultural product orders at other nodes in the circulation relationship chain of this batch of agricultural products. If it is detected that there is complaint data of other categories of agricultural product orders at a certain node, mark the corresponding user of this node with a risk and increase the sampling ratio of this user; if there is no complaint data of other categories of agricultural product orders at other nodes in the circulation relationship chain of this batch of agricultural products, increase the sampling ratio of the corresponding category of this batch of agricultural products.

[0051] In this implementation plan, the method for setting the spoilage probability threshold is as follows: Collect a large amount of spoilage probability data during the transportation and sales of agricultural products in the past, analyze the distribution of spoilage probabilities of agricultural products under different categories and transportation conditions, and determine a reasonable threshold. For example, through analysis, it is found that the spoilage probability of a certain type of agricultural product generally does not exceed 5% under normal transportation and storage conditions, then this value can be used as the initial spoilage probability threshold; reference can also be made to relevant agricultural product quality and safety standards, industry norms, and expert suggestions, and the threshold can be determined in combination with the actual situation. For example, some agricultural product industry associations may stipulate the allowable spoilage range of specific agricultural products within a certain transportation and storage period, and the spoilage probability threshold can be set accordingly; considering the costs of handling complaints and conducting traceability investigations, as well as the losses brought to consumers and enterprises due to the spoilage of agricultural products, a balance point can also be determined through cost-benefit analysis and used as the spoilage probability threshold. If the spoilage probability exceeds this threshold and the cost for the enterprise to take traceability and handling measures is less than the losses that may be caused by the spoilage of agricultural products, then this threshold is considered reasonable; by experimentally simulating the transportation and storage processes of agricultural products under different conditions, observing the spoilage situation of agricultural products, and determining the spoilage probabilities in different situations, the threshold can be set based on this. For example, in the laboratory, simulate different transportation environments such as different durations, temperatures, and humidities, and observe the spoilage speed and probability of agricultural products, so as to determine a spoilage probability threshold that conforms to the actual situation.

[0052] By comparing the spoilage probability with the threshold to trigger the problem traceability process and conducting targeted processing according to different situations, it is possible to accurately locate the links and nodes where agricultural product quality problems may occur, which helps to quickly find the root cause of the problem and take timely measures to solve the problem, ensuring the quality and safety of agricultural products; Dynamically adjusting the sampling ratio according to the complaint data and traceability results, and increasing the sampling intensity for node users or agricultural product categories at risk, can effectively increase the probability of detecting potential quality problems, timely discover and handle problem agricultural products, reduce the quality and safety risks of agricultural products, and at the same time can also prompt each node user to pay more attention to the quality of agricultural products; Timely handling of complaint data and taking corresponding measures helps to reduce the probability of consumers purchasing problem agricultural products, protect the legitimate rights and interests of consumers, improve consumers' trust in the quality of agricultural products, and maintain market order.

[0053] In summary, this application has at least the following effects:

[0054] By monitoring and tracing the whole process of agricultural products from production to sales, quality problems can be detected in a timely manner, high-risk batches can be marked and warned, the risk of problem agricultural products entering the market can be reduced, and the food safety of consumers can be guaranteed; construct a relationship chain for the circulation of agricultural products. Once a quality problem occurs, it can be quickly traced back to specific responsible parties such as farmers, suppliers, and sales points, facilitating the implementation of responsibilities and the investigation and improvement of problems; determine the sampling ratio according to the categories and production factors of agricultural products, conduct differential multi-index sampling, so that regulatory resources can be more reasonably allocated and regulatory efficiency can be improved; provide consumers with information such as the spoilage probability of agricultural products, enabling consumers to understand the product quality situation, enhancing consumer confidence, and at the same time prompting producers and operators to pay more attention to product quality; adjust the sampling strategy according to complaint data, which can continuously optimize the regulatory method and improve the effectiveness and pertinence of the monitoring of agricultural product quality and safety.

[0055] Those skilled in the art should understand that the embodiments of the present invention can be provided as methods. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0056] The present invention is described with reference to the flowchart of the method according to the embodiments of the present invention. It should be understood that the combination of each process in the flowchart can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in one Figure 1 process or multiple processes.

[0057] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means implements the functions specified in one Figure 1 process or multiple processes.

[0058] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one Figure 1 process or multiple processes.

[0059] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made to these embodiments by those skilled in the art once they learn the basic creative concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications that fall within the scope of the present invention.

[0060] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A method for monitoring, tracing and tracking the quality and safety of agricultural products, characterized in that, It includes the following steps: Step S1: Construct an agricultural product circulation relationship chain, determine the sampling ratio based on the agricultural product category and production factors, and then conduct differentiated multi-index sampling inspections on the agricultural products entered by farmers in the agricultural product circulation relationship chain; Step S2: For the agricultural products that pass the differentiated multi-index sampling inspection, predict the remaining safe storage time of the agricultural products based on the safe storage time prediction model, and then mark and give early warnings to high-risk batches; Step S3: Receive the agricultural product order demand, then identify the user's delivery address and agricultural product transportation end information, combine the user's delivery address, agricultural product transportation end information and the predicted remaining safe storage time of the agricultural products to predict the probability of agricultural product deterioration, and feedback the probability of agricultural product deterioration to the user waiting to place an order; Step S4: Judge whether to trigger providing an optimized order placement strategy for the user waiting to place an order based on the probability of agricultural product deterioration. The optimized order placement strategy includes delivery address optimization, transportation end optimization and agricultural product optimization; Step S5: Receive the complaint data of the user who placed the order, then determine the problem investigation and improvement requirements according to the traceability query of the complaint data, and adjust the sampling inspection strategy.

2. The method for monitoring, tracing and tracking the quality and safety of agricultural products according to claim 1, wherein Each node in the agricultural product circulation relationship chain includes farmers, suppliers, sales points, transportation ends and users who place orders.

3. A method for monitoring, tracing and tracking the quality and safety of agricultural products according to claim 1, characterized in that, The specific analysis of Step S1 is as follows: Identify the agricultural product category entered by farmers, then identify the category risk coefficient based on the agricultural product category, identify the production risk coefficient based on production factors, and then use the category risk coefficient and production risk coefficient to determine the sampling ratio; Determine multi-dimensional sampling inspection indicators according to the characteristics of the agricultural product category, then randomly select corresponding numbers of samples from the agricultural product batches entered by farmers according to the determined sampling ratio, and judge whether the agricultural products can enter the subsequent circulation link according to the test results of the selected samples.

4. A method for monitoring, tracing and tracking the quality and safety of agricultural products according to claim 1, characterized in that, The specific analysis of Step S2 is as follows: Obtain the agricultural product characteristic data based on the test of the selected samples. The agricultural product characteristic data includes basic attributes, physicochemical properties and physiological properties; Identify the transportation end data in the agricultural product circulation relationship chain. The transportation end data includes transportation equipment information and transportation environment information; Take the agricultural product characteristic data and transportation end data as the input of the safe storage time prediction model, output the remaining safe storage time of the agricultural products, compare the remaining safe storage time of the agricultural products with the safe storage time threshold. When the predicted remaining safe storage time of the agricultural products is lower than the safe storage time threshold, mark this batch of agricultural products as a high-risk batch.

5. A method for monitoring, tracing and tracking the quality and safety of agricultural products according to claim 1, characterized in that, The specific analysis of Step S3 is as follows: Use the order receiving port to receive the agricultural product purchase order to be placed by the user waiting to place an order in real time, and extract the delivery address information filled in by the user from the order details; Identify the agricultural product transportation end information related to this order according to the constructed agricultural product circulation relationship. The transportation end information includes the name of the logistics enterprise responsible for this transportation, transportation method, warehousing environment, expected transportation route and duration, and extract the geographical feature data in the expected transportation route; Call the predicted remaining safe storage time of this batch of agricultural products, and then output the deterioration probability value of this batch of agricultural products during the transportation to the user's delivery address according to the user's delivery address, agricultural product transportation end information and the predicted remaining safe storage time of the agricultural products.

6. A method for monitoring, tracing and tracking the quality and safety of agricultural products according to claim 1, characterized in that, The specific analysis of Step S4 is as follows: When the spoilage probability is greater than or equal to the spoilage probability threshold, trigger to provide an optimized order placement strategy for the user to be ordered. The optimization of the delivery address is specifically as follows: Retrieve the historical order records of the user to be ordered, extract the historical delivery address information of the user to be ordered from the historical order records, output the spoilage probability values respectively based on the historical delivery address information of the user to be ordered, mark the historical delivery addresses with spoilage probabilities lower than the spoilage probability threshold as optimized recommended addresses, and recommend them to the user to be ordered; The optimization of the transportation end is specifically as follows: Retrieve the remaining transportation methods and storage environments uploaded by the transportation end according to the constructed agricultural product circulation relationship, and then adjust the corresponding estimated transportation routes and durations, combine them to form an optimized set of transportation end information, output the spoilage probability values respectively based on the optimized set of transportation end information, mark the transportation end information with spoilage probabilities lower than the spoilage probability threshold as optimized recommended transportation end information, and recommend it to the user to be ordered; The optimization of the agricultural product is specifically as follows: Identify the agricultural product to be ordered by the user to be ordered, and then retrieve the similar agricultural products of this agricultural product, generate the corresponding spoilage probability values based on the similar agricultural products, mark the similar agricultural products with spoilage probabilities lower than the spoilage probability threshold as optimized recommended agricultural products, and recommend them to the user to be ordered.

7. The method for monitoring, tracing and tracking the quality and safety of agricultural products according to claim 7, characterized in that, Step S5 is specifically analyzed as follows: When detecting the complaint data of the user who has placed an order, check the spoilage probability feedback when the user who has placed an order ordered the agricultural product. When the spoilage probability is greater than or equal to the spoilage probability threshold, mark this user who has placed an order as a suspected malicious order placer, and trigger the artificial customer service to comfort the emotions of the user who has placed an order; When the spoilage probability is lower than the spoilage probability threshold, trigger the problem tracing process, retrieve the sales records of the same batch of agricultural products, and check whether there is other user's complaint data. If there is other user's complaint data, respectively detect whether there is other category of agricultural product order complaint data at other nodes in the agricultural product circulation relationship chain of this batch of agricultural products. If it is detected that there is other category of agricultural product order complaint data at a certain node, then mark the corresponding user at this node with a risk, and increase the sampling ratio of this user; If there is no other category of agricultural product order complaint data at other nodes in the agricultural product circulation relationship chain of this batch of agricultural products, then increase the sampling ratio of the corresponding category of this batch of agricultural products.

Images