Quality evaluation method and system for cigarette cut tobacco production process

Abstract

The invention discloses a quality evaluation method and system for a cigarette cut tobacco production process. The method comprises the steps of construction of a cut tobacco production quality evaluation index system, weight distribution of evaluation indexes, calculation of positive and negative ideal solutions, weighted calculation of quality scores and the like. The system comprises an evaluation index system construction module, a cut tobacco quality weight calculation module, a quality evaluation analysis module and an early warning module. According to the method, the steady-state data and the unsteady-state data are effectively distinguished in the tobacco shred making process and are included in the quality evaluation system, practical application shows that the method and the system can visually reflect the change condition between the quality of the current batch and the previous processing quality, the working procedure with the quality problem or the quality fluctuation is found, and the quality evaluation efficiency is improved. Weak links are fed back in time, and real-time evaluation reference and improvement suggestions are provided for operators and managers.

Description

Technical Field

[0001] The present invention relates to the field of cigarette cut tobacco processing, and particularly relates to a method and a system for evaluating the quality of the cigarette cut tobacco process. Background Art

[0002] Cut tobacco processing is the main link in cigarette production, which has the characteristics of many processing procedures, long process flow, and complex influencing factors, and has an important impact on the comprehensive quality of cigarettes, cost reduction and efficiency increase, and homogenized processing. Improving the process quality level of the cut tobacco process has been a long-term focus in the industry. Therefore, the evaluation method of the cut tobacco process quality has also become a current research hotspot.

[0003] A suitable quality evaluation method has practical significance for the control of the cut tobacco production process and helps to improve the stability of the cut tobacco process. Traditional cut tobacco quality evaluation methods include the conformity assessment method, the statistical assessment method, the sigma level method, etc. For example, Qi Lin et al. used the product of the sugar-alkali ratio and potassium content in the tobacco sample as the characteristic value, and based on the normal distribution test of the characteristic value and the calculation of the tobacco mixing uniformity, explored the quality stability and its change law of the tobacco in each cut tobacco process. Fan Shengxing et al. used statistical techniques to divide each quality index into multiple grades, determined the boundaries and scores of each grade, and carried out quality ranking according to the process score; in the "Cigarette Manufacturing Process Capability Evaluation Guidelines", the quality level of each key quality characteristic is converted into the number of defects per million opportunities, and the overall sigma level of the process is calculated by the geometric mean method. On the basis of these methods, Zhang Xinfeng used the network analysis method to determine the influence weight of the key cut tobacco processes on the cut tobacco quality, and then carried out quality evaluation through the calculation and ranking results of the weighted supermatrix; Luo Zhixue et al. used the QI index to construct a batch quality evaluation model for the whole cut tobacco processing process; Chen Deli et al. established a quality index characterization function based on the fuzzy algorithm and calculated the batch comprehensive score by the weighted method for quality evaluation. These methods have respectively proposed algorithm models for cut tobacco quality evaluation from different focuses, but for the daily process quality control and decision-making of cigarette factories, there are still deficiencies such as incomplete or unreasonable evaluation indicators for the whole cut tobacco process, the evaluation results are not sufficient to support real-time regulation, and the batch-to-batch comparability in the production process is relatively weak.

[0004] Therefore, the present invention combines the weighted TOPSIS method with the analytic hierarchy process, and proposes a method for evaluating the quality of the cut tobacco process, in order to comprehensively analyze the control level of the cut tobacco processing process and provide a scientific decision-making basis for improving the batch-to-batch quality stability. Summary of the Invention

[0005] In view of the deficiencies of the current method for evaluating the quality of cigarette cut tobacco production, such as emphasizing results over processes, having unintuitive conclusions, and lacking comparability, the present invention provides a method and system for evaluating the quality of the cigarette cut tobacco production process. This method is applicable to the longitudinal comparison of the quality of the cut tobacco production process with historical levels, can quickly identify weak batch indicators, and is conducive to promoting the continuous improvement of the quality of cut tobacco production.

[0006] The technical solution adopted by the present invention to solve its technical problems is as follows: A method for evaluating the quality of the cigarette cut tobacco production process, including S1 Constructing an index system for evaluating the quality of cut tobacco production Starting from quality characteristics, process parameters, and equipment performance levels, important evaluation elements affecting the quality of cut tobacco production are selected according to the hierarchical structure of work sections, processes, and process data, and an index system for evaluating the quality of cut tobacco production is established; S11 Data collection rules The collected process data includes quality characteristics, process parameters, and equipment parameters; among them, the quality characteristics include inlet and outlet moisture, the process parameters include flow rate type, opening type, and temperature type, and the equipment parameters are derived from the data statistically collected during the operation of the production equipment; A total of 50 batches of process data are collected for each process, and each index is collected once every 30 seconds for each batch; S12 Selection of quality evaluation indicators and evaluation statistics Appropriate evaluation statistics are selected to measure each key quality evaluation indicator of cut tobacco production, and the evaluation indicators for the cut tobacco production process are established according to the following criteria for the collected process data: ① For quality characteristics, CPK is used as the evaluation statistic: ; ; ; Among them, CPK is used to reflect whether the production process can reliably manufacture products that meet quality requirements within the specification limits; USL is the upper specification limit; μ is the process average; σ is the process standard deviation; LSL is the lower specification limit; ② For process parameters, the coefficient of variation (CV) is used as the evaluation statistic for the flow rate type, and the mean square error (MSE) can be used as the evaluation statistic for the opening type and temperature type: ; ; Among them, σ is the standard deviation of the sample, and μ is the average of the sample; is the i-th observed value, is the set target value, and n is the number of samples; ③For equipment parameters, which are mainly the data statistically collected during the operation of production equipment, the mean squared error (MSE) is used as the evaluation statistic; S2 Evaluation Index Weight Allocation Invite experts in the silk reeling production process to give an important judgment matrix for pairwise comparison of each evaluation index according to the ratio scale by scoring, and calculate the weights based on the analytic hierarchy process (AHP) of the geometric mean composite judgment matrix; Use the square root method to calculate the maximum eigenvalue , eigenvector and consistency ratio CR of each judgment matrix, and conduct a consistency test; The consistency test method is as follows: ; ; where n is the dimension of the judgment matrix; RI can be obtained by looking up the table; if , it is considered that the judgment matrix has satisfactory consistency; Average the weights of all experts who pass the consistency test to obtain the final weight values of each quality evaluation index; S3 Calculation of Positive and Negative Ideal Solutions The positive and negative ideal solutions respectively refer to the situations where the values of each evaluation statistic reach the optimal and worst at the same time, and are obtained through the statistical analysis of the quality data of the current batch to be evaluated and historical batches; Suppose there are i evaluation statistics of quality evaluation indexes in a silk reeling process, and the nearest n batches are selected as the evaluation reference. These n batches and the batch to be evaluated together form an evaluation matrix with n + 1 rows and i columns. The data in the j-th row and k-th column is the evaluation statistic of the k-th quality evaluation index of the j-th batch, denoted by ; Normalize the evaluation matrix. If the statistic of a quality evaluation index is of the extremely large type, then ; if the statistic of a quality evaluation index is of the extremely small type, then , thus forming the following matrix: ; The positive ideal solution can be formed by the maximum value of each column: ; The negative ideal solution is formed by the minimum value: , ; S4 Weighted Calculation of Quality Scores Calculate the distances between each batch and the positive and negative ideal solutions respectively: ; ; Among them, is the weight of the k-th quality evaluation index, ; Calculate the proximity of each batch to the ideal solution: , ; The larger it is, the higher the quality level of this batch; For the proximity of the (n + 1)-th batch to be evaluated, if it is lower than the average level of proximity , it indicates that the quality level of this batch is lower than the recent average level; S5 Intuitive evaluation of the quality of the silk-making process Convert into a value on a hundred-point scale, with the positive ideal solution (proximity is 1) being 100 points, and the average value of proximity being 80. Thus, the quality score of the (n + 1)-th batch to be evaluated can be calculated as: ; From this, a better dynamic longitudinal comparison of batch quality (comparison of data at different times in the same process) can be carried out. If the score is low, the possible problems should be analyzed, and corresponding quality improvement activities should be implemented if necessary.

[0007] The present invention also adopts the following technical solutions to solve its technical problems: A quality evaluation system for the cigarette silk-making process, including An evaluation index system construction module: According to the actual process flow of the production brand, count each section and different processes, and formulate the quality evaluation indexes and their evaluation statistics for each process; A silk-making quality weight calculation module: For the constructed processes, establish different numbers of quality evaluation indexes, give the weight of each index, and obtain the weight values of each quality evaluation index; A quality evaluation analysis module: Use the collected batch data as the evaluation reference, select the batch to be evaluated, and count the evaluation statistics of each quality evaluation index and the corresponding positive and negative ideal solutions; An early warning module: When the comprehensive quality score of the batch to be evaluated is low, the system starts to give an early warning, and the quality problems can be quickly discovered by combining the quality scores of each process, and can be improved and eliminated in time.

[0008] The beneficial effects brought by the present invention are: The present invention studies a method for effectively distinguishing between steady-state and non-steady-state data in the silk reeling process and incorporating them into the quality evaluation system. Then, based on the TOPSIS method, a quality evaluation method for silk reeling batches is constructed. The quality weights of each process and section are determined by the AHP method, and finally a systematic comprehensive evaluation system for the quality of silk reeling batches is formed. Practical applications show that this method and system can more intuitively reflect the changes between the quality of the current batch and the previous processing quality, identify processes with quality problems or quality fluctuations, timely feedback weak links, and provide real-time evaluation references and improvement suggestions for operators and managers. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0010] Figure 1 It is a schematic flowchart of the method according to the first embodiment of the present invention; Figure 2 It is a schematic diagram of the data analysis of Batch No. 6 according to the application embodiment of the present invention; Figure 3 It is a schematic diagram of the data analysis of Batch No. 10 according to the application embodiment of the present invention; Figure 4 It is a schematic diagram of the system framework according to the second embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0011] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the embodiments and the accompanying drawings. It should be understood that the general description above and the following detailed description are only exemplary and explanatory, and are not used to limit the present invention. The overview of various implementations or examples of the technologies described in the present invention is not a full disclosure of the entire scope of the disclosed technologies or all features. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0012] Unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the ordinary meanings understood by those of ordinary skill in the field to which the present invention pertains. The "first", "second" and similar terms used in the present invention do not denote any order, quantity or importance, but are only used to distinguish different components. Words such as "comprising" or "including" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. Words such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Upper", "lower", "left", "right", etc. are only used to represent relative positional relationships, and when the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0013] To keep the following description of the embodiments of the present invention clear and concise, the detailed descriptions of known functions and known modules and components are omitted in the present invention.

[0014] Referring to Figure 1 , the first embodiment of the present invention relates to a method for evaluating the quality of the cigarette cut tobacco process, including S1 Constructing a cut tobacco quality evaluation index system To construct a reasonable comprehensive cut tobacco quality evaluation index system, the key step is to scientifically select the important evaluation factors that affect the cut tobacco quality.

[0015] In this embodiment, starting from the quality characteristics, process parameters and equipment performance levels, following the principles of scientificity, systematicness, dynamics, operability and the combination of qualitative and quantitative, a comprehensive cut tobacco quality evaluation index system is established according to the hierarchical structure of the work section, process and process parameters; S11 Data collection rules The collected process data includes quality characteristics, process parameters and equipment parameters; among them, the quality characteristics include the inlet and outlet moisture, the process parameters include flow rate type, opening type and temperature type, and the equipment parameters are derived from the data statistically obtained during the operation of the production equipment; A total of 50 batches of process data are collected for each process, and each index (except the process duration) is collected once every 30 seconds for each batch; taking the thin plate drying process as an example, the collection method is shown in Table 1: Table 1 Process data collection form

[0016] S12 Selection of quality evaluation indexes and evaluation statistics Select appropriate evaluation statistics to measure each key cut tobacco quality evaluation index, and establish the evaluation indexes of the cut tobacco process for the collected process data according to the following criteria: ①For quality characteristics, since there are clear quality requirements in actual production, CPK is used as the evaluation statistic in this embodiment. CPK reflects whether the production process can reliably manufacture products that meet quality requirements within the specification limits. By measuring the centralization and dispersion of the process, it monitors and ensures the consistency and quality stability of products. The calculation method is as follows: ; ; ; where USL is the upper specification limit; μ is the process average; σ is the process standard deviation; LSL is the lower specification limit; ②For process parameters, for flow rate types (such as material flow rate), the coefficient of variation (CV) is used as the evaluation statistic. CV reflects the relative variability of the data and is suitable for evaluating the consistency and control level of the flow rate. For opening types (openings of various valves) and temperature types, the mean squared error (MSE) can be used as the evaluation statistic. MSE eliminates the directionality (positive or negative) of the error through squared differences and pays more attention to the magnitude of the error, which is especially suitable for applications that require precise control, such as valve opening and temperature control. The calculation method is as follows: ; ; where σ is the standard deviation of the sample, μ is the average of the sample; is the i-th observed value, is the set target value, and n is the number of samples; ③For equipment parameters, they are mainly the data statistically collected during the operation of production equipment. Taking the thin plate drying and flavoring process as an example, the opening of the inlet thin plate steam film valve and the frequency of the exhaust fan are the equipment parameters of this process. Since equipment parameters usually have clear standard requirements, MSE can also be used as the evaluation statistic.

[0017] S2 Evaluation Index Weight Allocation Invite experts in the cigarette making production process to give the pairwise comparison importance judgment matrix of each evaluation index by scoring according to the ratio scale shown in Table 2, and calculate the weights based on the analytic hierarchy process (AHP) of the geometric mean composite judgment matrix; Table 2 Ratio Scale Table

[0018] Note: The reverse ratio is used to represent the reverse importance comparison. For example, if the importance of factor A to factor B is 3, then the importance of factor B to factor A is 1 / 3.

[0019] Logical errors may occur when constructing the judgment matrix, so consistency checking is required. In this embodiment, the root method is used to calculate the maximum eigenvalue , eigenvector, and consistency ratio CR of each judgment matrix, and consistency checking is performed; the consistency checking method is as follows: ; ; where n is the dimension of the judgment matrix; RI can be obtained by looking up the table; if , it is considered that the judgment matrix has satisfactory consistency; Average the weights of all experts who pass the consistency check to obtain the final weight values of each quality evaluation index; S3 Calculation of positive and negative ideal solutions The positive and negative ideal solutions respectively refer to the situations where the values of each evaluation statistic reach the best and worst at the same time, and are obtained through statistical analysis of the quality data of the current batch to be evaluated and historical batches; Suppose there are evaluation statistics of i quality evaluation indexes in a silk reeling process. To avoid the influence of the change of sample data participating in the evaluation on the evaluation result, the method of large samples can be used to select the nearest n batches (usually ) as the evaluation reference. These n batches and the batch to be evaluated together form an evaluation matrix with n + 1 rows and i columns, and the following example data form in Table 3 is obtained. The data in the j-th row and k-th column is the evaluation statistic of the k-th quality evaluation index of the j-th batch, represented by ; Table 3 TOPSIS data form

[0020] Normalize the evaluation matrix. If the statistic of a quality evaluation index is of the extremely large type, then ; if the statistic of a quality evaluation index is of the extremely small type, then , thus forming the following matrix: ; The positive ideal solution can be formed by the maximum value of each column: ; The negative ideal solution is formed by the minimum value: , ; S4 Weighted calculation of quality scores Calculate the distances between each batch and the positive and negative ideal solutions respectively: ; ; Among them, is the weight of the k-th quality evaluation index, ; Calculate the proximity of each batch to the ideal solution: , ; The larger the , the higher the quality level of this batch; For the proximity of the (n + 1)-th batch to be evaluated, if it is lower than the average level of proximity , it indicates that the quality level of this batch is lower than the recent average level; it is necessary to analyze whether there are problems through the 5M1E analysis method (Man, Machine, Material, Method, Measurement, Environment) and solve them in a timely manner.

[0021] Intuitive evaluation of the quality of the S5 silk-making process In order to evaluate the quality of the silk-making process more intuitively, can be converted into a value on a 100-point scale, with the positive ideal solution (proximity is 1) being 100 points and the average value of proximity being 80. Then, the quality score of the (n + 1)-th batch to be evaluated can be calculated as: ; Thus, better dynamic longitudinal comparison of batch quality (comparison of data at different times in the same process) can be carried out. If the score is low, possible problems should be analyzed and corresponding quality improvement activities should be implemented if necessary.

[0022] The following uses a specific case to introduce the application of the cigarette silk-making process quality evaluation method described in the first embodiment.

[0023] According to the actual technological process of the D brand production enterprise, the silk-making quality is divided into 4 sections: tobacco sheet treatment, tobacco stem pretreatment, making and storing stem silk, and making, blending, and storing silk. Each section contains 2 - 8 processes with varying quantities respectively.

[0024] Taking the section of making, blending, and storing silk as an example, the quality evaluation indexes and their evaluation statistics of each process are shown in Table 4.

[0025] Table 4 Evaluation statistics table for the section of making, blending, and storing silk

[0026] Taking the thin plate drying process as an example, as can be seen from Table 4, there are 6 quality evaluation indexes in this process. Invite 5 experts to give the importance judgment matrix of pairwise comparison of each quality evaluation index. Table 5 shows an example of the judgment matrix given by one expert.

[0027] Comparison and Judgment Matrix of Quality Evaluation Indexes for the HT Thin Plate Cut Tobacco Drying Process

[0028] The maximum eigenvalue of the judgment matrix given by 5 experts was calculated respectively using the eigenvalue method and the weight vector, and a consistency test was carried out, as shown in Table 6

[0029] Table 6 Consistency Test Results

[0030] The weights of the 5 experts who passed the consistency test were averaged to obtain the final weight values of each quality evaluation index, as shown in the 7th column of Table 7

[0031] Table 7 Weight Vector

[0032] Using the data of 40 batches of D brand produced in June 2024 collected by the cut tobacco central control system as the evaluation reference, that is, n = 40. Taking the No. 1 batch produced in July 2024 as the batch to be evaluated, the evaluation statistics and corresponding positive and negative ideal solutions of each quality evaluation index are shown in Table 8

[0033] Table 8 Quality Evaluation Data for the HT Thin Plate Cut Tobacco Drying Process

[0034] As can be seen from Table 8, the opening of the steam film valve for the No. 1 batch is the negative ideal solution of the production data of 41 batches, representing that the control effect of the opening of the steam film valve for the No. 1 batch is the worst compared with the 40 historical control batches; while the evaluation statistics of the other 5 quality evaluation indexes are all between the positive and negative ideal solutions, and no obvious abnormality is shown. Substituting the weight and evaluation statistic data of each quality evaluation index into the calculation formula, the closeness of the No. 1 batch is obtained , and then the average closeness is obtained after calculating the closeness of all 40 control batches , and then the comprehensive quality score of the batch to be evaluated is calculated .

[0035] Taking the data of these 40 historical batches as the evaluation reference, taking the data of another 9 batches of D brand produced in July 2024 (represented by No. 2 - No. 10) as the evaluation objects, and calculating according to the same method, the closeness and comprehensive quality scores of 5 batches are obtained, as shown in Table 9

[0036] Table 9 Comprehensive Quality Scores of Evaluation Batches

[0037] It can be found in Table 9 that there are 3 batches (No.5, No.6 and No.10) with comprehensive quality scores below 80 points. The main reason for the low score of batch No.5 is that the overall process quality control level of the batch is not high, and the quality score is lower than 80 points. The relevant operators need to optimize the control model and strengthen management; the reasons for the low scores of batches No.6 and No.10 are the low scores of the loosening and drying processes. Draw radar charts for the scores of the quality evaluation indicators of each process, such as Figure 2 , Figure 3 shown.

[0038] Combination Figure 2 , Figure 3 Compared with the setting of process parameters of batches No.6 and No.10 in actual processing, the reasons for their low scores are as follows: (1) The scores of equipment parameters and some process parameters of batch No. 6 were both higher than 80 points, exceeding the average level of historical control batches; the scores of quality characteristic indicators and inlet water flow were low, and the weight of the latter was only 0.033, ranking 6th in the loosening and rehydration process. Therefore, it is believed that the low quality score of the loosening and rehydration process of batch No. 6 is mainly caused by quality characteristics. This may be due to the fluctuation of moisture content in the initial stage of production, which in turn caused the subsequent process to be difficult to stably control. It is recommended that relevant operators give priority to checking the control strategy and strengthening management.

[0039] (2) The process parameters and equipment parameter scores of batch No. 10 were not significantly abnormal, but the scores of quality characteristic indicators were low. The outlet moisture score was only 71.4 points, and the weight of this indicator in the leaf drying process was the largest at 0.35, which seriously affected the process quality score. After calculation, it was found that the outlet moisture CPK of this batch was as high as 11.33, which was significantly higher than the normal range. It is recommended that the relevant operators give priority to eliminating the equipment failure of the drying machine.

[0040] It can be seen that the silk-making process quality comprehensive evaluation method of this embodiment can be used for the batch comprehensive quality of the silk-making process, the evaluation result is consistent with the actual situation, and the method and model are effective. When the batch comprehensive quality score is low, the quality problems can be quickly discovered in combination with the quality scores of each process, and they can be improved and eliminated in time, which is an effective silk production process quality management method.

[0041] The second embodiment of the present invention relates to a cigarette shred process quality evaluation system, referring to Figure 4 ,include Evaluation index system construction module: According to the actual process flow of the production brand, statistics are collected for each section and different processes, and quality evaluation indicators and evaluation statistics for each process are formulated; Silk reeling quality weight calculation module: For the constructed processes, different numbers of quality evaluation indicators are established, the weight of each indicator is given, and the weight values of each quality evaluation indicator are obtained. Quality evaluation and analysis module: Using the collected batch data as the evaluation reference, the batches to be evaluated are selected, and the evaluation statistics and corresponding positive and negative ideal solutions of each quality evaluation indicator are counted. Early warning module: When the comprehensive quality score of the batch to be evaluated is low, the system starts to give an early warning, and it can quickly discover quality problems in combination with the quality scores of each process, and improve and eliminate them in a timely manner.

[0042] In the above embodiments, the descriptions of each embodiment have their own emphases. For the parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0043] Those of ordinary skill in the art can realize that the modules and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0044] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of each embodiment of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for evaluating the quality of cigarette shredded tobacco, characterized in that: include S1 Establishing a silk-making quality evaluation index system According to the hierarchical structure of work section, process and technological data, important evaluation factors affecting silk making quality are selected to establish a silk making quality evaluation index system; S11 Data Collection Rules The collected process data include quality characteristics, process parameters and equipment parameters; among which, quality characteristics include inlet and outlet moisture, process parameters include flow, opening and temperature, and equipment parameters are derived from the statistical data collected during the operation of production equipment; S12 Quality Evaluation Index and Selection of Evaluation Statistics Select appropriate evaluation statistics to measure each key quality evaluation index of silk making, and establish evaluation indicators of silk making process according to the following standards for the collected process data: ① For quality characteristics, CPK is used as the evaluation statistic; ② For process parameters, the coefficient of variation is used as the evaluation statistic for flow type, and the mean square error is used as the evaluation statistic for opening type and temperature type; ③For equipment parameters, the mean square error is used as the evaluation statistic; S2 evaluation index weight distribution By scoring, an important judgment matrix for pairwise comparison of each quality evaluation index is given, and the weights are calculated based on the hierarchical analysis method of the geometric mean composite judgment matrix; The square root method is used to calculate the maximum eigenvalue of each judgment matrix. , eigenvector and consistency ratio CR, and perform consistency check; if , then the judgment matrix is ​​considered to have satisfactory consistency; The weights of all experts who have passed the consistency test are averaged to obtain the final weight values ​​of each quality evaluation index; Calculation of positive and negative ideal solutions of S3 The positive and negative ideal solutions refer to the situations where the values ​​of each evaluation statistic reach the optimal and the worst at the same time, respectively, and are obtained through statistical analysis of the quality data of the current batch to be evaluated and the historical batches; Assume that a silk-making process has evaluation statistics of i quality evaluation indicators, and select the latest n batches as evaluation references. These n batches and the batch to be evaluated together form an evaluation matrix of n + 1 rows and i columns, where the data in the jth row and kth column is the evaluation statistic of the kth quality evaluation indicator of the jth batch, expressed as express; Normalize the evaluation matrix. If the statistic of a quality evaluation index is extremely large, then ; If the statistic of a quality evaluation index is extremely small, then , thus forming the following matrix: ； The maximum values ​​of each column can form a positive ideal solution: ； Construct a negative ideal solution from the minimum values: ， ； Weighted calculation of S4 quality score Calculate the distance between each batch and the positive and negative ideal solutions respectively: ； ； in, is the weight of the kth quality evaluation index, ; Compute how close each batch is to the ideal solution: , ; The larger it is, the higher the quality level of the batch; The closeness of the n+1th batch to be evaluated , if it is lower than the average level of proximity , it means that the quality level of this batch is lower than the recent average level.

2. The method for evaluating the quality of cigarette shredded tobacco according to claim 1, characterized in that: The method also includes S5 intuitive evaluation of the quality of the silk making process Will Converted to a percentage value, with the ideal solution as 100 points, the average value of the closeness is 80, so the quality score of the n + 1th batch to be evaluated can be calculated as: ； This allows for better dynamic longitudinal comparison of batch quality.

3. The method for evaluating the quality of cigarette shredded tobacco according to claim 1, characterized in that: S12 ① For quality characteristics, CPK is used as the evaluation statistic: ； ； ； Among them, CPK is used to reflect whether the production process can reliably produce products that meet quality requirements within the specification limits; USL is the upper specification limit; μ is the process mean; σ is the process standard deviation; LSL is the lower specification limit; ② For process parameters, the coefficient of variation CV is used as the evaluation statistic for flow type, and the mean square error MSE can be used as the evaluation statistic for opening type and temperature type: ； ； Among them, σ is the standard deviation of the sample, and μ is the mean of the sample; is the ith observation, is the set target value, n is the number of samples; ③For equipment parameters, which are mainly statistical data collected during the operation of production equipment, the mean square error (MSE) is used as the evaluation statistic.

4. The quality evaluation system of cigarette making process is characterized by: The system is used to apply the cigarette shred process quality evaluation method according to any one of claims 1 to 3, comprising: Evaluation index system construction module: According to the actual process flow of the production brand, statistics are collected for each section and different processes, and quality evaluation indicators and evaluation statistics for each process are formulated; Silk quality weight calculation module: for the constructed process, establish different numbers of quality evaluation indicators, give the weight of each indicator, and obtain the weight value of each quality evaluation indicator; Quality evaluation analysis module: using the collected batch data as an evaluation reference, selecting the batch to be evaluated, and calculating the evaluation statistics of each quality evaluation index and the corresponding positive and negative ideal solutions; Early warning module: When the comprehensive quality score of the batch to be evaluated is low, the system will start to warn. It can quickly discover quality problems based on the quality scores of each process and improve and eliminate them in time.

Images