A product traceability system based on big data mining

Abstract

The present invention relates to the field of product management, and discloses a product traceability system based on big data mining, comprising: a product data acquisition module, which collects product data based on an Internet of Things device and a barcode scanning device; a product data storage module, which is used to receive the product data and store it in a database as traceability data; a product traceability module, which processes raw material data, production process data and logistics data to obtain raw material scoring values, production process scoring values ​​and logistics transportation scoring values, and then compares them with their corresponding system thresholds, and judges whether there are abnormal items according to the comparison structure, and if so, it is determined to be the source of product problems, otherwise it is judged that the traceability fails; a traceability data verification module, which processes and analyzes the stored traceability data based on big data, obtains a quality trend coefficient, compares the calculated quality trend coefficient with a quality trend interval preset by the system, and judges the authenticity of the traceability data according to the comparison result.

Description

Technical Field

[0001] The present invention relates to the field of product management, and in particular to a product tracing system based on big data mining. Background Art

[0002] The purpose of product traceability is to track and record the entire life cycle of products from raw materials to final consumers, so that companies and consumers can understand the information of each process link such as the source, processing, transportation and sales of products, so as to improve product quality and safety and enhance consumer confidence.

[0003] Existing product traceability basically relies on suppliers to provide accurate data, which may contain false or incomplete information, and lacks verification of the provided information throughout the product life cycle. Combining the above two situations, when actually conducting product traceability, it may lead to traceability failure or unreliable traceability results. Summary of the invention

[0004] The purpose of the present invention is to provide a product traceability system based on big data mining to solve the above technical problems.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A product traceability system based on big data mining, comprising:

[0007] A product data collection module, which collects product data based on IoT devices and barcode scanning devices; the product data includes raw material data, production process data, and logistics data;

[0008] A product data storage module, used to receive the product data and store it in a database as traceability data;

[0009] The product traceability module is used to trace products with quality problems. By processing raw material data, production process data and logistics data, the raw material score, production process score and logistics transportation score are obtained, and then compared with the corresponding system thresholds. According to the comparison structure, it is determined whether there are abnormal items. If so, it is determined to be the source of the product problem, otherwise it is determined that the traceability fails;

[0010] The traceability data verification module processes and analyzes the stored traceability data based on big data, obtains the quality trend coefficient, compares the calculated quality trend coefficient with the quality trend interval preset by the system, and determines the authenticity of the traceability data based on the comparison results.

[0011] The present invention uses a product traceability module to trace the source of products with quality problems. By processing raw material data, production process data and logistics data, the raw material score value, the production process score value and the logistics transportation score value are obtained, and then compared with the corresponding system thresholds. According to the comparison structure, it is judged whether there is an abnormal item. If so, it is determined to be the source of the product problem, otherwise it is judged that the traceability fails. Under the premise that the product has quality problems and the source of the product problem is not found, it is first judged that there is a problem with the product traceability data, and then the quality problem that occurs after the product reaches the consumer is judged. Therefore, the traceability data is verified by the traceability data verification module, and the quality trend coefficient is calculated, and the calculated quality trend coefficient is compared with the quality trend interval preset by the system. If the quality trend coefficient exceeds the quality trend interval, the traceability data is judged to be false, and the source of the problem of the same batch of products is determined to be the product processing process. If the quality trend coefficient is within the quality trend interval, it is judged that the quality trend of the same batch of products is consistent with the actual situation. Therefore, the traceability data of the product is true, and the source of the quality problem of the product is judged to be the consumer itself. If the quality trend coefficient is lower than the quality trend interval, it is judged that the calculated time span is too small to truly reflect the quality trend changes of the product. Therefore, the current time span is doubled and calculated again until the authenticity of the current product traceability data is determined. Through the above process, the source of the problem can be accurately judged during the traceability process to avoid misjudgment and traceability failure, and potential traceability data fraud problems can be discovered.

[0012] Furthermore, the working process of the product traceability module is as follows:

[0013] The raw material score w1, production process score w2 and logistics and transportation score w3 are respectively compared with their corresponding system thresholds w 1th 、w 2th 、w 3th Make a comparison;

[0014] If w1≥w 1th or w2 ≥ w 2th or w3 ≥ w 3th If at least one of the following conditions is met, the process corresponding to the product is determined to be the source of the problem;

[0015] Otherwise, the overall quality score K of the current product is calculated based on the raw material score w1, the production process score w2 and the logistics and transportation score w3; where: Where N is the total number of products with quality problems in the same batch, W i1 , W i2 , W i3 They are the reference values ​​of raw material score, production process score, and logistics and transportation score obtained based on big data;

[0016] The calculated overall quality score K and the system warning value K th0 Make comparisons;

[0017] If K ≥ K th0 , then it is judged that there are problems in the entire process of the current product;

[0018] Otherwise, the authenticity of the traceability data is further determined through the traceability data verification module.

[0019] Furthermore, the process of obtaining the raw material score w1, the production process score w2 and the logistics and transportation score w3 is as follows:

[0020] According to formula (1): The raw material score w1 is calculated; where q is the dimensionless coefficient, A is the raw material source grade of the product, B is the pesticide residue in the raw material, and C is the quality grade; θ, is the preset proportional coefficient;

[0021] According to formula (2): Calculate and obtain the production process score w2; where D is the cleanliness level of the product's production workshop, E is the quality inspection level of the product after processing, and s is the preset proportional coefficient;

[0022] The logistics transportation score w3 is calculated by formula (3): w3 = (F2-F0) × β1 + (F0-F1) × β2; where F1~F2 is the allowable logistics transportation time range, F0 is the actual logistics transportation time, and β1 and β2 are preset proportional coefficients.

[0023] Furthermore, the process of judging the authenticity of traceability data is as follows:

[0024] The quality trend coefficient X is calculated by the formula: X = P1×ɑ1+P2×ɑ2+P3×ɑ3;

[0025] Where P1 is the raw material trend coefficient, P2 is the production process trend coefficient, P3 is the logistics trend coefficient; ɑ1, ɑ2, w3 are weight coefficients;

[0026] The quality trend coefficient X and the system preset quality trend interval [X f , X z ] for comparison;

[0027] If X>X z And K<K th0 , then the traceability data is judged to be false;

[0028] If X∈[X f , X z ] and K<K th0 , then the traceability data is judged to be true;

[0029] If X<X f And K<K th0 , then expand the sampling period ta~tb by 2 times and repeat the above process.

[0030] Furthermore, through formula (4):

[0031]

[0032] According to formula (5):

[0033]

[0034] According to formula (6):

[0035]

[0036] Calculate and obtain the raw material trend coefficient P1, production process trend coefficient P2 and logistics trend coefficient P3;

[0037] Among them, v is the reference coefficient, which is comprehensively formulated based on experimental data and historical data, ω k is the influence coefficient, w 1max (t) is the maximum value of the raw material score during the period ta to tb, w 10 (t) is the curve of raw material score value preset by the system over time, w 20 (t) is the curve of the production process score value of the system intervention over time, w 30 (t) is the system preset logistics transportation score value changing with time curve, ta~tb is the sampling time period, and n is the total number of sampling time periods.

[0038] Furthermore, the system also includes an early warning module, which is used to issue an early warning based on the result of judging the authenticity of the traceability data, that is, when the traceability data is judged to be false, an early warning signal is sent to the management personnel of the system background.

[0039] Beneficial effects of the present invention:

[0040] The present invention uses a product traceability module to trace the source of products with quality problems. By processing raw material data, production process data and logistics data, the raw material score value, the production process score value and the logistics transportation score value are obtained, and then compared with the corresponding system thresholds. According to the comparison structure, it is judged whether there is an abnormal item. If so, it is determined to be the source of the product problem, otherwise it is judged that the traceability fails. Under the premise that the product has quality problems and the source of the product problem is not found, it is first judged that there is a problem with the product traceability data, and then the quality problem that occurs after the product reaches the consumer is judged. Therefore, the traceability data is verified by the traceability data verification module, and the quality trend coefficient is calculated, and the calculated quality trend coefficient is compared with the quality trend interval preset by the system. If the quality trend coefficient exceeds the quality trend interval, the traceability data is judged to be false, and the source of the problem of the same batch of products is determined to be the product processing process. If the quality trend coefficient is within the quality trend interval, it is judged that the quality trend of the same batch of products is consistent with the actual situation. Therefore, the traceability data of the product is true, and the source of the quality problem of the product is judged to be the consumer itself. If the quality trend coefficient is lower than the quality trend interval, it is judged that the calculated time span is too small to truly reflect the quality trend changes of the product. Therefore, the current time span is doubled and calculated again until the authenticity of the current product traceability data is determined. Through the above process, the source of the problem can be accurately judged during the traceability process to avoid misjudgment and traceability failure, and potential traceability data fraud problems can be discovered. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The present invention will be further described below in conjunction with the accompanying drawings.

[0042] Figure 1 It is a schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] See also Figure 1 As shown, the present invention is a product tracing system based on big data mining, comprising:

[0045] A product data collection module, which collects product data based on IoT devices and barcode scanning devices; the product data includes raw material data, production process data, and logistics data;

[0046] A product data storage module, used to receive the product data and store it in a database as traceability data;

[0047] The product traceability module is used to trace products with quality problems. By processing raw material data, production process data and logistics data, the raw material score, production process score and logistics transportation score are obtained, and then compared with the corresponding system thresholds. According to the comparison structure, it is determined whether there are abnormal items. If so, it is determined to be the source of the product problem, otherwise it is determined that the traceability fails;

[0048] The traceability data verification module processes and analyzes the stored traceability data based on big data, obtains the quality trend coefficient, compares the calculated quality trend coefficient with the quality trend interval preset by the system, and determines the authenticity of the traceability data based on the comparison results.

[0049] The process of judging the authenticity of traceability data is as follows:

[0050] The quality trend coefficient X is calculated by the formula: X = P1×ɑ1+P2×ɑ2+P3×ɑ3;

[0051] Where P1 is the raw material trend coefficient, P2 is the production process trend coefficient, P3 is the logistics trend coefficient; ɑ1, ɑ2, w3 are weight coefficients;

[0052] The quality trend coefficient X and the system preset quality trend interval [X f , X z ] for comparison;

[0053] If X>X z And K<K th0 , then the traceability data is judged to be false;

[0054] If X∈[X f , X z ] and K<K th0 , then the traceability data is judged to be true;

[0055] If X<X f And K<K th0 , then expand the sampling period ta~tb by 2 times and repeat the above process.

[0056] The present invention uses a product traceability module to trace the source of products with quality problems. By processing raw material data, production process data and logistics data, the raw material score value, the production process score value and the logistics transportation score value are obtained, and then compared with the corresponding system thresholds. According to the comparison structure, it is judged whether there is an abnormal item. If so, it is determined to be the source of the product problem, otherwise it is judged that the traceability fails. Under the premise that the product has quality problems and the source of the product problem is not found, it is first judged that there is a problem with the product traceability data, and then the quality problem that occurs after the product reaches the consumer is judged. Therefore, the traceability data is verified by the traceability data verification module, and the quality trend coefficient is calculated, and the calculated quality trend coefficient is compared with the quality trend interval preset by the system. If the quality trend coefficient exceeds the quality trend interval, the traceability data is judged to be false, and the source of the problem of the same batch of products is determined to be the product processing process. If the quality trend coefficient is within the quality trend interval, it is judged that the quality trend of the same batch of products is consistent with the actual situation. Therefore, the traceability data of the product is true, and the source of the quality problem of the product is judged to be the consumer itself. If the quality trend coefficient is lower than the quality trend interval, it is judged that the calculated time span is too small to truly reflect the quality trend changes of the product. Therefore, the current time span is doubled and calculated again until the authenticity of the current product traceability data is determined. Through the above process, the source of the problem can be accurately judged during the traceability process to avoid misjudgment and traceability failure, and potential traceability data fraud problems can be discovered.

[0057] The working process of the product traceability module is as follows:

[0058] The raw material score w1, production process score w2 and logistics and transportation score w3 are respectively compared with their corresponding system thresholds w 1th 、w 2th 、w 3th Make a comparison;

[0059] If w1≥w 1th or w2 ≥ w 2th or w3 ≥ w 3th If at least one of the following conditions is met, the process corresponding to the product is determined to be the source of the problem;

[0060] Otherwise, the overall quality score K of the current product is calculated based on the raw material score w1, the production process score w2 and the logistics and transportation score w3; where:

[0061] Where N is the total number of products with quality problems in the same batch, W i1 , W i2 , W i3 They are the reference values ​​of raw material score, production process score, and logistics and transportation score obtained based on big data;

[0062] The calculated overall quality score K and the system warning value K th0 Make comparisons;

[0063] If K ≥ K th0 , then it is judged that there are problems in the entire process of the current product;

[0064] Otherwise, the authenticity of the traceability data is further determined through the traceability data verification module.

[0065] The present invention adopts the formula Calculate the overall quality score K of the product with quality problems. Obviously, if |w1-w 1th |、|w1-w 1th |、|w3-w 3th The larger the value, the greater the overall quality score K of the product. The greater the overall quality volatility of the product, the greater the probability of quality problems. The calculated overall quality score K and the system warning value K th0 Compare; if K ≥ K th0 , it is judged that there are problems in the entire process of the current product. Otherwise, the authenticity of the traceability data is further judged through the traceability data verification module; thus, the source of product problems can be preliminarily analyzed from the fluctuation of the overall quality score to eliminate traceability errors.

[0066] The process of obtaining the raw material score w1, the production process score w2, and the logistics and transportation score w3 is as follows:

[0067] According to formula (1): The raw material score w1 is calculated; where q is the dimensionless coefficient, A is the raw material source grade of the product, B is the pesticide residue in the raw material, and C is the quality grade; θ, is the preset proportional coefficient;

[0068] According to formula (2): Calculate and obtain the production process score w2; where D is the cleanliness level of the product's production workshop, E is the quality inspection level of the product after processing, and s is the preset proportional coefficient;

[0069] The logistics transportation score w3 is calculated by formula (3): w3 = (F2-F0) × β1 + (F0-F1) × β2; where F1~F2 is the allowable logistics transportation time range, F0 is the actual logistics transportation time, and β1 and β2 are preset proportional coefficients.

[0070] In the present invention, the scoring value of the corresponding process is obtained by calculating each process of the product, thereby reflecting the actual situation of each process of the current product.

[0071] According to formula (4):

[0072]

[0073] According to formula (5):

[0074]

[0075] According to formula (6):

[0076]

[0077] Calculate and obtain the raw material trend coefficient P1, production process trend coefficient P2 and logistics trend coefficient P3;

[0078] Among them, v is the reference coefficient, which is comprehensively formulated based on experimental data and historical data, ω k is the influence coefficient, w 1max (t) is the maximum value of the raw material score during the period ta to tb, w 10 (t) is the curve of raw material score value preset by the system over time, w 20 (t) is the curve of the production process score value of the system intervention over time, w 30 (t) is the system preset logistics transportation score value changing with time curve, ta~tb is the sampling time period, and n is the total number of sampling time periods.

[0079] The present invention provides a method for obtaining a raw material trend coefficient P1, a production process trend coefficient P2 and a logistics trend coefficient P3. Specifically, the method comprehensively considers the deviation between the raw material score value variation data w1(t) of the current product over multiple sampling time periods ta~tb and the system preset raw material score value variation data over time, the deviation between the production process score value variation data w2(t) of the current product over the sampling time period ta~tb and the preset curve, and the deviation between the logistics transportation score value variation curve of the current product over the sampling time period ta~tb and the preset curve, and obtains the quality trend coefficient X by calculating the formula: =P1×ɑ1+P2×ɑ2+P3×ɑ3, so as to accurately reflect the quality trend of a single process of the product and the quality trend of the entire process.

[0080] The system also includes an early warning module, which is used to issue an early warning based on the result of judging the authenticity of the traceability data, that is, when the traceability data is judged to be false, an early warning signal is sent to the management personnel of the system background.

[0081] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A product traceability system based on big data mining, characterized in that: include: A product data collection module, which collects product data based on IoT devices and barcode scanning devices; the product data includes raw material data, production process data, and logistics data; A product data storage module, used to receive the product data and store it in a database as traceability data; The product traceability module is used to trace products with quality problems. By processing raw material data, production process data and logistics data, the raw material score, production process score and logistics transportation score are obtained, and then compared with the corresponding system thresholds. According to the comparison structure, it is determined whether there are abnormal items. If so, it is determined to be the source of the product problem, otherwise it is determined that the traceability fails; The traceability data verification module processes and analyzes the stored traceability data based on big data to obtain the quality trend coefficient, compares the calculated quality trend coefficient with the quality trend interval preset by the system, and determines the authenticity of the traceability data based on the comparison results; The working process of the product traceability module is as follows: The raw material score w1, production process score w2 and logistics and transportation score w3 are respectively compared with their corresponding system thresholds w 1th 、w 2th 、w 3th Make a comparison; If w1≥w 1th or w2 ≥ w 2th or w3 ≥ w 3th If at least one of the following conditions is met, the process corresponding to the product is determined to be the source of the problem; Otherwise, the overall quality score K of the current product is calculated based on the raw material score w1, the production process score w2 and the logistics and transportation score w3; where: Where N is the total number of products with quality problems in the same batch, W i1 , W i2 , W i3 They are the reference values ​​of raw material score, production process score, and logistics and transportation score obtained based on big data; The calculated overall quality score K and the system warning value K th0 Make comparisons; If K ≥ K th0 , then it is judged that there are problems in the entire process of the current product; Otherwise, the authenticity of the traceability data is further determined through the traceability data verification module; The process of obtaining the raw material score w1, the production process score w2, and the logistics and transportation score w3 is as follows: According to formula (1): The raw material score w1 is calculated; where q is the dimensionless coefficient, A is the raw material source grade of the product, B is the pesticide residue in the raw material, and C is the quality grade; θ, is the preset proportional coefficient; According to formula (2): Calculate and obtain the production process score w2; where D is the cleanliness level of the product's production workshop, E is the quality inspection level of the product after processing, and s is the preset proportional coefficient; The logistics transportation score w3 is calculated by formula (3): w3 = (F2-F0) × β1 + (F0-F1) × β2; where F1-F2 is the range of allowed logistics transportation time, F0 is the actual logistics transportation time, and β1 and β2 are preset proportional coefficients; The process of judging the authenticity of traceability data is as follows: The quality trend coefficient X is calculated by the formula: X = P1×ɑ1+P2×ɑ2+P3×ɑ3; Where P1 is the raw material trend coefficient, P2 is the production process trend coefficient, P3 is the logistics trend coefficient; ɑ1, ɑ2, w3 are weight coefficients; The quality trend coefficient X and the system preset quality trend interval [X f , X z ] for comparison; If X>X z And K<K th0 , then the traceability data is judged to be false; If X∈[X f , X z ] and K<K th0 , then the traceability data is judged to be true; If X<X f And K<K th0 , then expand the sampling period ta~tb by 2 times and repeat the above process; According to formula (4): According to formula (5): According to formula (6): Calculate and obtain the raw material trend coefficient P1, production process trend coefficient P2 and logistics trend coefficient P3; Among them, v is the reference coefficient, which is comprehensively formulated based on experimental data and historical data, ω k is the influence coefficient, w 1max (t) is the maximum value of the raw material score during the period ta to tb, w 10 (t) is the curve of raw material score value preset by the system over time, w 20 (t) is the curve of the production process score value of the system intervention over time, w 30 (t) is the system preset logistics transportation score value changing with time curve, ta~tb is the sampling time period, and n is the total number of sampling time periods.

2. The product traceability system based on big data mining according to claim 1 is characterized in that: The system also includes an early warning module, which is used to issue an early warning based on the result of judging the authenticity of the traceability data, that is, when the traceability data is judged to be false, an early warning signal is sent to the management personnel of the system background.

Images