A TOPSIS-based method for evaluating the overall quality of yarn production

Abstract

This invention discloses a comprehensive quality evaluation method for yarn processing based on TOPSIS. The main design concept of this invention lies in combining the weighted TOPSIS method with the analytic hierarchy process (AHP). A quality evaluation index system for the yarn processing process is established according to a hierarchical structure of sections, processes, and process parameters. AHP is then used for weight allocation. By measuring various quality characteristic indicators, the batch quality and quality of each yarn processing step in the overall yarn processing process are evaluated. This invention can intuitively reflect the changes in current batch quality compared to previous processing quality. It is suitable for longitudinal comparison and evaluation of the yarn processing process against its own historical quality level, identifying processes or process parameters with quality problems or declining quality levels. It provides timely feedback on weak links in the yarn production process, offering real-time evaluation references and improvement suggestions for operators and managers, which is conducive to promoting continuous improvement of batch quality and enhancing the overall quality level of the yarn processing process.

Description

Technical Field

[0001] This invention relates to the field of cigarette manufacturing, and in particular to a comprehensive quality assessment method for cigarette processing based on TOPSIS. Background Technology

[0002] The tobacco processing stage is a major part of cigarette production. It is characterized by many processing steps, a long process flow, and complex influencing factors. Therefore, it is a key process for improving the intrinsic quality of cigarettes, reducing material and energy consumption, and reducing tar and harm.

[0003] Improving the quality of the yarn-making process has long been a focus of attention in the industry, and therefore, the evaluation methods for yarn-making process quality have become a current research hotspot. Appropriate quality evaluation strategies can effectively promote scientific decision-making in process parameter control and improve the stability of yarn-making process quality.

[0004] Traditional tobacco processing quality evaluation mechanisms in this industry mainly include conformity assessment, statistical assessment, and sigma level methods. For example, the overall standard deviation of a process section can be calculated using process standard deviation to evaluate the quality of the tobacco processing process; or statistical techniques can be used to subdivide various quality indicators into multiple levels, determine the boundaries and scores of each level, and rank the quality based on the process scores; furthermore, in the cigarette manufacturing process capability assessment guidelines, the quality level of each critical quality characteristic is converted into defects per million opportunities (DPMO), and the geometric mean method is used to calculate the overall sigma level of the process.

[0005] Based on these traditional approaches, some proposals have been put forward to use network analysis to determine the impact weights of key silk-making processes on silk quality, and then to evaluate quality by calculating and ranking the results of weighted hypermatrix; or to use the QI index to construct a batch quality evaluation model for the entire silk-making process; or to establish a quality index representation function based on fuzzy algorithms and to evaluate quality by calculating the comprehensive batch score using a weighted method.

[0006] Although these strategies propose algorithmic mechanisms for evaluating silk production quality from different perspectives, they still have shortcomings such as incomplete evaluation indicators, emphasis on results over process, lack of intuitiveness in evaluation results, and lack of comparability. Summary of the Invention

[0007] In view of the above, the present invention aims to provide a comprehensive quality evaluation method for yarn production based on TOPSIS, so as to achieve a comprehensive analysis of the control level of the yarn production process, reduce quality fluctuations between batches, and thus ensure the stability of process quality.

[0008] The technical solution adopted in this invention is as follows:

[0009] This invention provides a method for evaluating the overall quality of yarn production based on TOPSIS, including:

[0010] Based on the production process, several silk-making sections are pre-divided, and each silk-making section includes multiple production processes.

[0011] Select several corresponding quality characteristic indicators in different production processes;

[0012] Based on different indicator types, the evaluation statistics for each of the quality characteristic indicators are determined; wherein, the indicator types include at least quality result type and process control parameter type;

[0013] The quality characteristic indicators are divided into steady-state data and non-steady-state data. Based on the steady-state data of the quality characteristic indicators and the preset proportional coefficient, the quality evaluation value of the quality characteristic indicators of the batch to be evaluated is obtained.

[0014] Based on the statistical data of quality evaluation values ​​of the quality characteristic indicators of the batch to be evaluated and several historical batches obtained within a preset time range, the positive and negative ideal solutions of the quality characteristic indicators are obtained.

[0015] The weights of each quality characteristic index are determined using the analytic hierarchy process (AHP).

[0016] Using the positive and negative ideal solutions, each of the quality evaluation values, and the corresponding weights, the quality evaluation results of the batch to be evaluated in each of the production processes and the comprehensive quality evaluation result that integrates the evaluation results of all production processes are obtained.

[0017] In at least one possible implementation, the determination of the evaluation statistics for each of the quality characteristic indicators includes:

[0018] The process capability index is used as an evaluation statistic for quality characteristic indicators of quality results.

[0019] Based on parameter characteristics and expert experience, the coefficient of variation and / or mean square error are used as evaluation statistics for quality characteristic indicators of process control parameters.

[0020] In at least one possible implementation, dividing the quality characteristic index into steady-state data and non-steady-state data includes:

[0021] The variance is calculated using the autocovariance and autocorrelation function of the autocorrelation process to define the steady-state range of the autocorrelation process.

[0022] In at least one possible implementation, obtaining the quality evaluation value of the quality characteristic index of the batch to be evaluated includes calculating the process capability index for quality characteristic indicators of quality outcome type according to the following formula:

[0023]

[0024] in, G represents the overall process capability evaluation value and the process capability evaluation value under steady-state conditions for the batch to be evaluated, respectively; B G S These represent the overall cumulative material data and the cumulative material data under steady state for the batch to be evaluated, respectively.

[0025] In at least one possible implementation, obtaining the positive and negative ideal solutions for the quality characteristic index includes:

[0026] Several historical batches were collected as reference samples;

[0027] An initial data matrix is ​​constructed from the quality evaluation values ​​of each quality characteristic index, which consists of the reference sample and the batch to be evaluated.

[0028] Based on the attributes of each of the quality characteristic indicators, the initial data matrix is ​​normalized to obtain the target matrix;

[0029] The maximum value of each column in the target matrix constitutes the positive ideal solution, and the minimum value of each column in the target matrix constitutes the negative ideal solution.

[0030] In at least one possible implementation, the step of using the analytic hierarchy process (AHP) to determine the weights of each quality characteristic index includes:

[0031] Each of the quality characteristic indicators is compared and scored pairwise based on the experience of a single expert.

[0032] Based on the scoring results and the preset scale, an importance judgment matrix is ​​generated for each of the quality characteristic indicators corresponding to the expert's experience.

[0033] Calculate the largest eigenvalue and the corresponding weight vector of the importance judgment matrix;

[0034] The consistency test was used to examine the importance judgment matrices obtained from the experience of different experts.

[0035] The mean of the weight vectors in all the importance judgment matrices that have passed the consistency test is calculated to obtain the weight value of each quality characteristic index.

[0036] In at least one possible implementation, obtaining the quality evaluation results of the batch to be evaluated in each of the production processes includes:

[0037] By combining the weight values ​​of each quality characteristic index with the corresponding quality evaluation values, the distance between each quality characteristic index and the positive and negative ideal solutions is calculated respectively.

[0038] Based on the distance, the closeness of the batch to be evaluated to the corresponding positive and negative ideal solutions in each of the production processes is calculated, and the closeness is used as the quality evaluation result.

[0039] In at least one possible implementation, the evaluation method further includes:

[0040] The proximity score is converted into a percentage value and combined with the preset average proximity score to obtain the quality assessment score of the batch to be evaluated.

[0041] The main design concept of this invention is to address the shortcomings of current methods for evaluating yarn quality, such as emphasizing results over process, lacking intuitive conclusions, and lacking comparability. It combines the weighted TOPSIS (Technique for Order Preference by Similarity to an Ideal Solution) method with the Analytic Hierarchy Process (AHP). A quality evaluation index system for the yarn manufacturing process is established according to a hierarchical structure of sections, processes, and process parameters. AHP is used for weight allocation. By measuring various quality characteristic indicators, the batch quality and quality of each yarn manufacturing process are evaluated. This invention can more intuitively reflect the changes in current batch quality compared to previous processing quality. It is suitable for longitudinal comparison and evaluation of the yarn manufacturing process against its own historical quality level, identifying processes or process parameters with quality problems or declining quality levels. It provides timely feedback on weak links in the yarn manufacturing process, offering real-time evaluation references and improvement suggestions for operators and managers, which is conducive to promoting continuous improvement of batch quality and enhancing the overall quality level of the yarn manufacturing process. Attached Figure Description

[0042] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described below with reference to the accompanying drawings, wherein:

[0043] Figure 1 A flowchart illustrating the overall framework of the TOPSIS-based comprehensive quality assessment method for yarn production provided in this embodiment of the invention.

[0044] Figure 2 This is a schematic diagram of the structure of the silk-making quality evaluation system provided in an embodiment of the present invention. Detailed Implementation

[0045] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0046] This invention proposes an embodiment of a comprehensive quality evaluation method for yarn production based on TOPSIS, specifically, as follows: Figure 1 As shown, it includes:

[0047] Step S1: Based on the production process, several silk-making sections are pre-divided, and each silk-making section includes multiple production processes;

[0048] Specifically, based on the actual production process of an enterprise, the quality of tobacco processing can be divided into four stages: tobacco sheet processing, tobacco stem pretreatment, stem and tobacco processing and storage, and tobacco processing, blending, and storage. Each stage contains two to eight different processes. (See reference...) Figure 2 As shown.

[0049] In order to comprehensively evaluate the quality of yarn production, based on the established standards for process quality assessment, step S2 is performed to select several corresponding quality characteristic indicators in different production processes.

[0050] Step S3: Determine the evaluation statistics for each quality characteristic indicator based on different indicator types; wherein, the indicator types can mainly include two types: quality results and process control parameters.

[0051] Among them, the quality characteristic indicators of quality results mainly include outlet moisture and temperature. These indicators usually have clear process specification requirements, so C can be used. pk As an evaluation statistic;

[0052] Process parameter-type quality characteristic indicators mainly include flow rate, opening degree, and air temperature. These indicators usually do not have explicit process specification requirements. Therefore, in actual operation, based on parameter characteristics and expert experience, the coefficient of variation (CV) can be used as the evaluation statistic for flow rate indicators (such as material flow rate, steam flow rate, feed flow rate, etc.), while the mean square error (MSE) can be used as the evaluation statistic for opening degree and air temperature. The calculation methods for the two can be found below:

[0053]

[0054] Where μ, σ, and T represent the mean, variance, and target value, respectively. Taking the stem and filament production section as an example, the quality characteristic indicators and their evaluation statistics for each production process involved in this section are shown in Table 1.

[0055] Table 1 Evaluation Statistics of the Straw Preparation and Storage Section

[0056]

[0057]

[0058] Step S4: Divide the quality characteristic indicators into steady-state data and non-steady-state data, and based on the steady-state data of the quality characteristic indicators and the preset proportional coefficient, calculate the quality evaluation value of each quality characteristic indicator of the batch to be evaluated.

[0059] Specifically, since most statistical indicators for measuring quality levels are based on a process in a steady state, such as process capability indices and process standard deviations, it is essential to distinguish between steady and unstable states before evaluating the quality of each quality characteristic indicator. An unstable state refers to a production process where quality indicators (such as moisture content and temperature) exceed the expected range or are in an unstable production state (including material start-up, material finish-up, and material interruption). According to traditional statistical quality control theory, a process is considered steady-state when its quality indicator data are stable within the range of μ ± 3σ (where μ is the process mean and σ is the process standard deviation); anything outside this range is considered unstable.

[0060] (I) Regarding the steady-state process data:

[0061] For the silk production process, due to the significant autocorrelation of process data, if the calculation method under the independent distribution case is used to calculate its variance, it will often be relatively small. When using the 3σ method to determine the steady state, it is usually relatively narrow, which makes the data that is actually in a steady state incorrectly judged as unsteady. This situation often occurs when using conventional control charts to monitor the silk production process.

[0062] In response to this, the present invention proposes that in order to more accurately define the steady state of an autocorrelation process, its variance must be calculated more accurately, which can be done by using the autocovariance and autocorrelation function of the autocorrelation process.

[0063] First, calculate the autocovariance function:

[0064]

[0065] Specifically, when k = 0:

[0066]

[0067] Next, calculate the autocorrelation function:

[0068]

[0069] The variance of the autocorrelation process can be expressed as:

[0070]

[0071] The range of the steady state can be obtained using the method of three standard deviations:

[0072]

[0073] (II) Regarding the non-steady-state process data: Because the quality level of the process is low and varies significantly when it is in a non-steady-state state, it is usually impossible to measure it using appropriate quality statistical indicators. Therefore, a proportionality coefficient mechanism can be used to calculate the overall quality of the batch to be evaluated. For example, the process capability index C for the moisture content at the outlet of a certain process can be used. pk :

[0074]

[0075] in, G represents the overall process capability evaluation value and the process capability evaluation value under steady-state conditions for the batch to be evaluated, respectively; B G S G represents the batch cumulative material flow rate and the steady-state cumulative material flow rate, respectively; the longer the non-steady-state time, the higher the G value. S / G B The smaller it is, the better. The smaller it is.

[0076] Therefore, by distinguishing between steady-state and non-steady-state data for each process, and by calculating the overall quality evaluation value of each quality characteristic within the process using a proportionality coefficient, both non-steady-state and steady-state data can be incorporated into the quality evaluation system.

[0077] Following the previous text, step S5 involves obtaining the positive and negative ideal solutions for the quality characteristic indicators based on the statistical data of the quality evaluation values ​​of the batch to be evaluated and historical batches.

[0078] Specifically, positive and negative ideal solutions refer to the simultaneous optimal and worst-case scenarios for each evaluation index, respectively. These can be obtained through statistical analysis of the quality data of the current batch to be evaluated and historical batches. Assuming a certain silk-making process has i quality characteristic indicators, to avoid the influence of changes in the sample data on the evaluation results, a large sample method can be used to select the most recent n batches (n should be large enough, usually n≥50) as evaluation references. These n batches and the batch to be evaluated together form an n+1 row, i column matrix, resulting in the data format shown in Table 2 below. The data in the j-th row and k-th column represents the quality evaluation value of the k-th quality characteristic indicator of the j-th batch, denoted by r. jk express.

[0079] Table 2 TOPSIS Data Format

[0080]

[0081] After normalizing the data in each column of Table 2, if a higher quality evaluation value for the indicator is better, then:

[0082]

[0083] If the smaller the indicator, the better, then:

[0084]

[0085] This results in the following matrix:

[0086]

[0087] The positive ideal solution can be constructed from the maximum values ​​of each column:

[0088] U + =[max(r′) j1 ), max(r′ j2 ), …max(r′) ji )], j = 1, 2, ..., n+1;

[0089] Similarly, the negative ideal solution can be constructed from the minimum values ​​of each column:

[0090] U - =[min(r′) j1 ), min(r′ j2 ), …min(r′ ji )).

[0091] Taking the stem expansion and drying process as an example, data from 96 batches of Grade D cigarettes produced from June to October 2021, collected by the cigarette factory's tobacco processing quality information system, were used as evaluation references, i.e., n=96. One batch produced in November was selected as the batch to be evaluated. The evaluation statistics of each quality characteristic index and the corresponding positive and negative ideal solutions are shown in Table 3.

[0092] Table 3 Quality evaluation data for the expansion and drying process of the fibrous stems

[0093]

[0094] In Table 3, the first row of the batch to be evaluated contains the original data of the evaluation statistics of each quality characteristic of the batch. The second row contains the data after processing the original data together with the data of 96 batches used as evaluation references using the normalization method above equation (8). The original data and normalized data of the reference batches are not described in detail. Similarly, the first and second rows of the positive and negative ideal solutions are also the original data and the normalized data, respectively.

[0095] Step S6: Use the analytic hierarchy process (AHP) to determine the weights of each quality characteristic index;

[0096] Taking the stem expansion and drying process mentioned earlier as an example, this section illustrates an exemplary method for obtaining weights using AHP. In Table 1 above, there are six quality characteristic indicators in the stem expansion and drying process. First, the importance judgment matrix for pairwise comparison of each quality characteristic indicator is given according to a preset scale of 1-9 through expert scoring, as shown in Table 4.

[0097] Table 4 Comparison and Judgment Matrix of Quality Characteristic Indicators for the Expansion and Drying Processes of Fiber Straw

[0098]

[0099] The maximum eigenvalue λ of the matrix is ​​calculated using the eigenvalue method. max The weight vector, consisting of the weights of each quality characteristic index, is shown in column 2 of Table 5. Since expert scoring is highly subjective, to more objectively evaluate the importance of each index, multiple experts can score separately. The maximum eigenvalue and weight vector of the judgment matrix obtained from the scores of the other experts are shown in columns 3-6 of Table 5.

[0100] Table 5 Weight Vectors and Maximum Eigenvalues

[0101]

[0102]

[0103] The random consistency index of the judgment matrices given by each expert is calculated based on the consistency discrimination method:

[0104]

[0105] Where t represents the number of quality characteristic indicators. For example, the CI of expert 1's judgment matrix is ​​0.0414, and RI can be obtained by looking up the random consistency index table. n＝6 =1.26, therefore the consistency ratio is C. R =CI / RI=0.033, usually in C R When the value is less than 0.1, the judgment matrix can be considered to have passed the consistency test.

[0106] The judgment matrices of all experts are tested using a similar method. If a matrix fails the test, the experts need to re-score the data or a new expert needs to be selected to re-score it. The final weight values ​​for each quality characteristic index are obtained by averaging the weights of all experts who passed the consistency test, as shown in column 7 of Table 5.

[0107] Step S7: Using the positive and negative ideal solutions, each of the quality evaluation values ​​and the corresponding weights, obtain the quality evaluation results of the batch to be evaluated in each process and the comprehensive quality evaluation result of the batch.

[0108] Based on the preceding text, this section presents a better method for calculating scores using weights, which can be used to calculate the distance between the quality characteristic indicators of each batch and the aforementioned positive and negative ideal solutions:

[0109]

[0110] Where, p k The weight of the k-th quality characteristic index,

[0111] Based on the distance, calculate the proximity of each batch to the ideal solution:

[0112]

[0113] Understandably, Q j The larger the value, the higher the quality level of the batch.

[0114] As for the proximity Q of the (n+1)th batch to be evaluated n+1 If it is lower than the average level obtained from the similarity of each batch. This indicates that the quality level of this batch is lower than the average level of this period (related to the historical batch selection range), and it is necessary to analyze whether there are any problems from the aspects of 5M1E and resolve them in a timely manner.

[0115] Furthermore, in order to more intuitively evaluate the quality of the yarn-making process, Q can be used... j Convert to a percentage system, for example, with a positive ideal solution (closeness of 1) as 100 points, the average closeness score. The score is 80, therefore the quality score F of the (n+1)th batch to be evaluated is... n+1 It can then be calculated as:

[0116]

[0117] This allows for better dynamic longitudinal comparison of batch quality (data comparison at different times in the same process). If the score is low, measures should be taken to analyze the problems and make corrections to achieve continuous improvement in process quality.

[0118] Based on the previous example, the following specific explanation is provided: The proximity Q of the batch to be evaluated can be calculated from equations (9) and (10). n+1 =0.705, then calculate the proximity of all batches to obtain the average proximity. The quality score F of the process in the batch to be evaluated is calculated using equation (11). n+1 =83.7.

[0119] Following a similar approach, using the same 96 historical batches as evaluation references, and the 10 batches produced in November (represented as No.1-10) as evaluation objects, the comprehensive quality scores of the 10 batches were calculated using the aforementioned weighting analysis method and the weighted TOPSIS quality comprehensive evaluation model, as shown in Table 6.

[0120] Table 6. Process quality scores and overall batch quality scores for each batch.

[0121]

[0122] Table 6 shows that two batches had an overall quality score below 80 (the set average score). The low score of batch No. 2 was mainly due to poor quality control in some key processes, especially the leaf drying process, which had the highest weighting and only scored 73.5 points. The investigation revealed that this was mainly due to condensation on the cylinder wall causing fluctuations in the outlet moisture content, and the operators were notified to eliminate this issue promptly. The low score of batch No. 6 was due to poor overall process quality control, with multiple processes scoring below 80. The investigation revealed that this was mainly due to a decrease in operator focus before shift handover, and the workshop was notified to strengthen management.

[0123] Therefore, the weighted TOPSIS quality comprehensive evaluation strategy proposed in this invention can effectively evaluate the overall batch quality and the quality of each process in the silk-making process. The evaluation results are consistent with the actual situation. Moreover, when the overall batch quality score is low, the quality problems can be quickly identified by combining the quality scores of each process, and timely improvements and eliminations can be made. It is undoubtedly an effective quality management method for the silk-making production process.

[0124] In summary, the main design concept of this invention is to address the shortcomings of current methods for evaluating yarn quality, such as emphasizing results over process, lacking intuitive conclusions, and lacking comparability. It combines the weighted TOPSIS method with the analytic hierarchy process (AHP) to establish a quality evaluation index system for the yarn production process according to a hierarchical structure of sections, processes, and process parameters. AHP is then used for weight allocation. By measuring various quality characteristic indicators, the batch quality and quality of each yarn production process are evaluated. This invention can more intuitively reflect the changes in current batch quality compared to previous processing quality. It is suitable for longitudinal comparison and evaluation of the yarn production process against its own historical quality levels, identifying processes or process parameters with quality problems or declining quality levels. It provides timely feedback on weak links in the yarn production process, offering real-time evaluation references and improvement suggestions for operators and managers, which is conducive to promoting continuous improvement of batch quality and enhancing the overall quality level of the yarn production process.

[0125] In this embodiment of the invention, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, A and B simultaneously, or B alone. A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects have an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one of a, b, and c can represent: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, and c can be single or multiple.

[0126] The above description of the structure, features, and effects of the present invention is based on the embodiments shown in the figures. However, the above are only preferred embodiments of the present invention. It should be noted that the technical features involved in the above embodiments and their preferred methods can be reasonably combined and matched by those skilled in the art to form a variety of equivalent solutions without departing from or changing the design concept and technical effects of the present invention. Therefore, the present invention is not limited to the scope of implementation shown in the figures. Any changes made in accordance with the concept of the present invention, or modifications to equivalent embodiments, that do not exceed the spirit covered by the specification and figures, should be within the protection scope of the present invention.

Claims

1. A comprehensive quality assessment method for yarn production based on TOPSIS, characterized in that, include: Based on the production process, several silk-making sections are pre-divided, and each silk-making section includes multiple production processes. Select several corresponding quality characteristic indicators in different production processes; Based on different indicator types, evaluation statistics for each quality characteristic indicator are determined; wherein, the indicator types include at least quality result type and process control parameter type; the quality characteristic indicators of the quality result type have clear process specification requirements, while the quality characteristic indicators of the process parameter type do not have clear process specification requirements. The determination of the evaluation statistics for each of the quality characteristic indicators specifically includes: using the process capability index as the evaluation statistics for quality characteristic indicators of the quality result category; and using the coefficient of variation and / or mean square error as the evaluation statistics for quality characteristic indicators of the process control parameter category, based on parameter characteristics and expert experience. The quality characteristic index is divided into steady-state data and non-steady-state data, specifically including: using the autocovariance and autocorrelation function of the autocorrelation process to calculate the variance in order to define the steady-state range of the autocorrelation process; Based on the steady-state data of the quality characteristic indicators and the preset proportional coefficient, the quality evaluation value of the quality characteristic indicators of the batch to be evaluated is obtained. Specifically, for quality characteristic indicators of quality results, the process capability index is calculated according to the following formula: ； in, , These represent the overall process capability evaluation value and the process capability evaluation value under steady-state conditions for the batch to be evaluated, respectively. , These represent the overall cumulative material data and the cumulative material data under steady state for the batch to be evaluated, respectively; and the longer the non-steady-state time, the greater G. S / G B The smaller it is, the better. The smaller it is; Based on the statistical data of quality evaluation values ​​of the quality characteristic indicators of the batch to be evaluated and several historical batches obtained within a preset time range, the positive and negative ideal solutions of the quality characteristic indicators are obtained. The weights of each quality characteristic index are determined using the analytic hierarchy process (AHP). Using the positive and negative ideal solutions, each of the quality evaluation values, and the corresponding weights, the quality evaluation results of the batch to be evaluated in each of the production processes and the comprehensive quality evaluation result integrating the evaluation results of all production processes are obtained. The process of obtaining the quality evaluation results of the batch to be evaluated in each of the production processes includes: The weight values ​​of each quality characteristic index are combined with the corresponding quality evaluation values ​​to calculate the distance between each quality characteristic index and the positive and negative ideal solutions. Based on the distance, the closeness of the batch to be evaluated to the corresponding positive and negative ideal solutions in each production process is obtained, and the closeness is used as the quality evaluation result.

2. The TOPSIS-based comprehensive quality evaluation method for yarn production according to claim 1, characterized in that, The process of obtaining the positive and negative ideal solutions for the quality characteristic index includes: Several historical batches were collected as reference samples; An initial data matrix is ​​constructed from the quality evaluation values ​​of each quality characteristic index, which consists of the reference sample and the batch to be evaluated. Based on the attributes of each of the quality characteristic indicators, the initial data matrix is ​​normalized to obtain the target matrix; The maximum value of each column in the target matrix constitutes the positive ideal solution, and the minimum value of each column in the target matrix constitutes the negative ideal solution.

3. The method for comprehensive quality evaluation of yarn production based on TOPSIS according to any one of claims 1 to 2, characterized in that, The method of using the analytic hierarchy process (AHP) to determine the weights of each quality characteristic index includes: Each of the quality characteristic indicators is compared and scored pairwise based on the experience of a single expert. Based on the scoring results and the preset scale, an importance judgment matrix is ​​generated for each of the quality characteristic indicators corresponding to the expert's experience. Calculate the largest eigenvalue and the corresponding weight vector of the importance judgment matrix; The consistency test was used to examine the importance judgment matrices obtained from the experience of different experts. The mean of the weight vectors in all the importance judgment matrices that have passed the consistency test is calculated to obtain the weight value of each quality characteristic index.

4. The TOPSIS-based comprehensive quality evaluation method for yarn production according to claim 1, characterized in that, The evaluation method also includes: The proximity score is converted into a percentage value and combined with the preset average proximity score to obtain the quality assessment score of the batch to be evaluated.

Images