CN101414183A - Cigarette working procedure quality overall evaluation system and method based on gray correlation analysis - Google Patents

Abstract

The invention discloses a comprehensive cigarette process quality evaluation system based on grey correlation analysis and a method thereof, aims at overcoming the disadvantage that the existing single process evaluation method can not comprehensively evaluate the overall process quality of the cigarette process, and changing the situation of fuzzy process capability evaluation of a whole line at present. The evaluation system and the method thereof help comprehensively evaluates the quality of a plurality of processes, and can effectively grasp comprehensive situation of the process quality so as to provide basis for the continuous improvement in the process quality, and provide assurance for quality improvement of cigarette products. The system and the method combine the single-process quality evaluation method and the grey correlation analysis method. The method comprises the following steps: firstly, computing process capability indexes of various process parameters; then, comprehensively evaluating a Cpk value by the grey correlation method; finally, obtaining a comprehensive evaluation result which is taken as a process quality rank of various evaluated objects. Production practice can be effectively directed by analyzing the Cpk and the process quality ranking result.

Description

Cigarette working procedure quality overall evaluation system and method thereof based on grey correlation analysis

Technical field

The present invention relates to a kind of cigarette working procedure quality overall evaluation system and method thereof, relate to tobacco working procedure quality overall evaluation field based on grey correlation analysis.

Background technology

Processing technology is as one of four key elements of cigarette product design, and the quality of cigarette product is had bigger influence.The cigarette processing technology has not been a simple pursuit physical index, has pursued the process that reduces consumption at present, but distillation is process and the technical method that improves the cigarette product inherent quality.Stable, good processing technology can keep the intrinsic fine quality of tobacco leaf to greatest extent, effectively remove the undesirable element of raw material itself, improve the maximum value of cigarette raw material, create have higher product, the cigarette product of peculiar fragrance style and taste feature.Along with raising, the especially proposition of " Chinese style cigarette " international, the domestic consumption level, China's cigarette processing technology is had higher requirement.

At present domestic control to process stability and product inherent quality with abroad compare, also have big gap.Therefore, need further to improve the technology level of China's cigarette processing, improve the control ability of technology, give full play to cigarette Material Characteristics and use value the cigarette product quality.That this just needs is objective, cigarette working procedure quality evaluation method accurately, so that grasp the actual conditions of process quality, for the raising of technology level provides foundation.Operation is the basic link that cigarette product forms, and process quality is subjected to the influence of a plurality of factors such as operating personnel, production environment, machinery and equipment.In the production run of product, production technology is carried out the process capability analysis, understanding and grasping process capability is a control and an important means that guarantees product quality.Process capability index is the index of quantitatively characterizing process capability level height, the expression technological level satisfies the degree of technological parameter code requirement, by calculating the process capability index of each technological parameter in the cigarette working procedure, can grasp the degree of each technological parameter conformance with standard standard.

China part enterprise has begun to carry out working procedure quality, and for example the requirement supplier at the beginning of 2006 of Haier carries out the process capability evaluation.Estimate and require to comprise: formulate the control chart of process chart, quality control planning, Critical to quality table, each operation Critical to quality, the process capability evaluation Cpk of critical process etc.At cigarette industry, Qingdao cigar mill utilizes the process capability of statistical Process Control analysis of cigarette machine weight control device.Present evaluation to cigarette working procedure is based on single operation substantially, promptly certain specific technological parameter is carried out statistical Process Control or process capability analysis.The major defect of single operation evaluation is to hold the overall process quality situation of sliver, for example cannot compare the process quality quality of sliver under different months.How to select scientific and effective working procedure quality overall evaluation method, change the situation of at present evaluation of sliver process capability being blured, thereby improve technological level, improve the quality of cigarette product, become tobacco enterprise problem anxious to be solved.

Summary of the invention

Purpose of the present invention is exactly the problem that exists in estimating at above-mentioned cigarette working procedure quality, has proposed a kind of cigarette working procedure quality overall evaluation system and method based on grey correlation analysis.At first the instrument detecting key process parameter that comes out is calculated its process capability index respectively, with the method for grey correlation analysis whole technology is carried out working procedure quality overall evaluation again, thereby overcome the shortcoming that single operation evaluation only is conceived to certain technological parameter, can hold the process quality situation of sliver objective, accurately, for updating of process quality provides foundation.

For achieving the above object, the present invention has adopted following technical scheme:

Based on the cigarette working procedure quality overall evaluation system of grey correlation analysis, comprise an authority management module, described authority management module is connected with the data importing module; Described data importing module is selected to be connected with pretreatment module with data, and data are selected with pretreatment module data to be carried out pre-service; Described data selection is connected with single working procedure quality module with pretreatment module; Described single working procedure quality module is connected with the working procedure quality overall evaluation module of grey correlation analysis; The working procedure quality overall evaluation module of described grey correlation analysis is showed with evaluation result and is connected with output module; Described data are selected also to select module with the operation kind with pretreatment module; Described single working procedure quality module also is connected with the technological standards database.

Described authority management module is connected with rights database and carries out data transfer, and the role in the system, user, user's group are defined and the management of authority separately.

Described data importing module is connected with the technological parameter database carries out data transfer, in the data importing system, and can add data, operation such as deletion, renewal, the real-time management data.

Described operation kind selection module is connected with operation kind template base carries out data transfer, selects corresponding working procedures from operation kind template base, and can create new operation kind.

Methods such as operation kind that described data are selected and the pretreatment module basis is selected or singular value, repetition values, missing values are carried out pre-service to data.

Described technological standards database is connected with the technological standards administration module and carries out data transfer; Described technological standards administration module provides preservation, operations such as modification.

Described evaluation result is showed and output module passes to output device with the result, intuitively shows result with the form of chart; Output device passes to the evaluation result database again with data.

A kind of cigarette working procedure quality overall evaluation method based on grey correlation analysis, this method may further comprise the steps:

Step1: the user inputs user name, password, role, and authority management module is according to the data search user right in the rights database.Satisfy condition, then allow this user further to operate, otherwise force this user to withdraw from native system;

Step2: the user by the data importing module with in the data importing system in the technological parameter database;

Step3: the operation kind that the user selects module will estimate from the selection of operation kind template base by the operation kind, and send the data to data selection and pretreatment module;

Step4: data are selected according to the operation kind data of import system to be carried out pre-service with pretreatment module;

Step5: the data in single working procedure quality module combined process standard database are carried out the evaluation of process capability index to each technology of pretreated cigarette working procedure;

Step6: the working procedure quality overall evaluation module of grey correlation analysis is carried out comprehensive evaluation to the technique process quality of cigarette working procedure;

Step7: evaluation result is showed to be connected with output module and evaluation result is sent to output device and is saved in the evaluation result database;

Among the described Step5 each technology of pretreated cigarette working procedure is carried out the evaluation of process capability index, may further comprise the steps:

(1): by the every technological parameter of instrument detecting cigarette, with technological parameter data recording, storage;

(2): reject the wrong or special sample in the technological parameter data;

(3): with the technological standards of every technological parameter is that standard center M and specification tolerance T store;

(4): technological parameter according to estimating the purpose grouping, for example will be compared the good and bad situation of different month process qualitys, and then the technological parameter in each month is as one group, and every group as an evaluation object;

(5): calculate the sample average x and the sample standard deviation σ of each each technological parameter of evaluation object respectively, computing formula is:

x＝∑x _i/n

$\mathrm{\σ}=\sqrt{\mathrm{\Σ}{(x_i-x)}^2/(n-1)}$

Wherein n is a number of samples in each evaluation object;

(6): with sample average x, sample variance σ, standard center M, upper specification limit T _U, specification lower limit T _LWith specification tolerance T substitution process capability index computing formula, calculate single process capability index X of each evaluation object _i(i=1,2, Λ, m), m is the number of evaluation object,

$C_{\mathrm{pk}}=\frac{|T_U-T_L|}{6\mathrm{\&sigma;}}(1-k)=\frac{\mathrm{<figure-callout\; id="6"\; label="T"\; filenames="A200810237698E00161.png"\; state="\{\{state\}\}">T</figure-callout>}}{6\mathrm{\&sigma;}}(1-k)$

Wherein $k=\frac{2(M-x)}{|T_U-T_L|}=\frac{2(M-x)}{T}$

Single process capability index X of each evaluation object _iThe row vector of forming by the Cpk of each technological parameter.

Technique process quality to cigarette working procedure among the described Step6 is carried out comprehensive evaluation, may further comprise the steps:

(1): with single process capability index X of each evaluation object _i(i=1,2, Λ, m) as with reference to ordered series of numbers, X _i(i=1,2, Λ, m) the capable vector formed of the maximal value of every row ordered series of numbers Y as a comparison, wherein m is the number of evaluation object;

(2): to reference sequence X _i(i=1,2, Λ m) carries out nondimensionalization with comparison ordered series of numbers Y with the initial value method and handles, and the ordered series of numbers after the processing is designated as respectively

With Y ＇

$X_i^{\&prime;}\left(k\right)=X_i\left(k\right)/Y\left(k\right)$

Y＇(k)＝Y(k)/Y(k)

Wherein k represents k technological parameter;

(3): ask

Each element and Y ＇ when the difference in prostatitis ${\mathrm{\&Delta;}}_i\left(k\right)=|X_i^{\&prime;}\left(k\right)-Y\&prime;\left(k\right)|$ (i=1,2, Λ, m), Δ _i(k) form matrix of differences;

(4): obtain the minimum value a and the maximal value b of matrix of differences, make resolution ratio ρ=0.5, compute associations coefficient r _i(k):

$r_i\left(k\right)=\frac{a+\mathrm{\&rho;b}}{{\mathrm{\&Delta;}}_i\left(k\right)+\mathrm{\&rho;b}}$

R wherein _i(k) expression

Degree of correlation with Y ＇ (k);

(5): ask grey relational grade γ _i(i=1,2, Λ, m).Obtain the integrated value of each evaluation object correlation coefficient:

${\mathrm{\&gamma;}}_i=\frac{1}{n}\underset{k}{\mathrm{\&Sigma;}}{\mathrm{\&gamma;}}_i\left(k\right)\mathrm{\&omega;}$

Wherein ω is the weight of the technological parameter of evaluation object;

(6): the grey relational grade to each evaluation object sorts, and obtains the quality ordering of the process quality integrated value of each evaluation object.

Each module functions is as follows:

Authority management module: the definition and the management of authority separately that provide role, user, user in the system to organize have guaranteed the secure access of database.Role Management is used for the role and the corresponding role-security of define system.User management is used for the user and the corresponding user right of define system.The user organizes management, is used for the user's group and the corresponding authority of define system;

The data importing module: this module comprises the import feature of data.The operation data that the import feature of data imports the user will to analyze with the user according to demand, thus data analyzed more targetedly and efficiently.This module also provide to data add, operation such as deletion, renewal, the real-time management data;

The operation kind is selected module: according to the operation that will estimate, the user selects corresponding working procedures from operation kind class libraries, also can create new operation kind; Also comprise the selection of evaluation method in this module, comprise monthly and estimating, estimate per year etc.;

Data are selected and pretreatment module: this module provides data to select and the data preprocessing function.Select the operation data that to analyze according to selected operation kind.This module also provides methods such as singular value, repetition values, missing values that data are carried out pre-service, to reject the wrong or special sample in the data;

Technological standards administration module: realize the function such as preservation, modification of technological standards data of each technological parameter of each operation, for single working procedure quality module provides support.

Single working procedure quality module: the function that realizes single working procedure quality;

The working procedure quality overall evaluation module of grey correlation analysis: the working procedure quality overall evaluation function that realizes the grey correlation method;

Evaluation result is showed and output module: the result of the working procedure quality overall evaluation of single process capability evaluation result and grey correlation analysis is represented with the situation of chart.The user analyzes evaluation result in conjunction with chart, and evaluation result and user's oneself analysis result can be outputed in the evaluation result database.

The system and method for cigarette working procedure comprehensive evaluation of the present invention, its core and gordian technique are: earlier each technological parameter of cigarette is carried out single process capability evaluation, promptly calculate the process capability index (Cpk) of each technological parameter respectively, the integration capability of cigarette working procedure quality is estimated from integral body with the method for grey correlation analysis then.At last from the good and bad ordering of the grey relational grade of each evaluation object, and, carry out interpretation of result in conjunction with the process capability index (Cpk) of each technological parameter.

Based on the cigarette working procedure quality overall evaluation system of grey correlation analysis, its advantage and beneficial effect are:

(1) overcome the shortcoming of only considering in the working procedure quality in the past from the part, method with grey correlation analysis is carried out comprehensive evaluation from integral body to process capability, can hold the quality situation of overall wire operation on the whole, solve at present to the fuzzy situation of sliver process capability evaluation.

(2) according to the comprehensive evaluation result of system to process quality, for the relatively poor object of process quality situation, can ascertain the reason from aspects such as production equipment, personnel operation, raw material, production environments as early as possible, promote the improvement of relevant function work, thereby promote the raising of technological level with this.For process quality situation object preferably, should continue to keep.

(3) to promoting enterprise's process management innovation, improve technology and process management level, performance raw material use value improves the directive significance that the cigarette inherent quality has reality.

Figure of description

Fig. 1 is that the module of cigarette working procedure quality evaluation system of the present invention is formed structural representation.

Fig. 2 is based on the process flow diagram of the cigarette working procedure quality overall evaluation method of grey correlation analysis for the present invention.

Wherein, 1, authority management module; 2, data importing module; 3, data are selected and pretreatment module; 4, the operation kind is selected module; 5, technological standards administration module; 6, single working procedure quality module; 7, the working procedure quality overall evaluation module of grey correlation analysis; 8, evaluation result is showed and output module; 9, rights database; 10, technological parameter database; 11, operation kind template base; 12, technological standards database; 13, output device; 14, evaluation result database.

Embodiment

The invention will be further described below in conjunction with accompanying drawing and embodiment.

In Fig. 1, based on the cigarette working procedure quality overall evaluation system of grey correlation analysis, comprise an authority management module 1, authority management module 1 is connected with data importing module 2; Data importing module 2 is selected to be connected with pretreatment module 3 with data, and data are selected to carry out pre-service with 3 pairs of data of pretreatment module; Data are selected to be connected with single working procedure quality module 6 with pretreatment module 3; Single working procedure quality module 6 is connected with the working procedure quality overall evaluation module 7 of grey correlation analysis; The working procedure quality overall evaluation module 7 of grey correlation analysis is showed with evaluation result and is connected with output module 8; Data are selected also to select module 4 with the operation kind with pretreatment module 3; Single working procedure quality module 6 also is connected with technological standards database 12.

Authority management module 1 is connected with rights database 9 and carries out data transfer, and the role in the system, user, user's group are defined and the management of authority separately.

Data importing module 2 is connected with technological parameter database 10 carries out data transfer, in the data importing system, and can add data, operation such as deletion, renewal, the real-time management data.

Operation kind selection module 4 is connected with operation kind template base 11 carries out data transfer, selects corresponding working procedures from operation kind template base 11, and can create new operation kind.

Methods such as operation kind that data are selected and pretreatment module 3 bases are selected or singular value, repetition values, missing values are carried out pre-service to data.

Technological standards database 12 is connected with technological standards administration module 5 and carries out data transfer; Described technological standards administration module 5 provides preservation, operations such as modification.

Evaluation result is showed and output module 8 passes to output device 13 with the result, intuitively shows result with the form of chart; Output device 13 passes to evaluation result database 14 again with data.

A kind of cigarette working procedure quality overall evaluation method based on grey correlation analysis, this method may further comprise the steps:

Step1: the user inputs user name, password, role, and authority management module 1 is according to the data search user right in the rights database 9.Satisfy condition, then allow this user further to operate, otherwise force this user to withdraw from native system;

Step2: the user by data importing module 2 with in the data importing system in the technological parameter database 10;

Step3: the operation kind that the user selects module 4 will estimate from 11 selections of operation kind template base by the operation kind, and send the data to data selection and pretreatment module 3;

Step4: data are selected according to the operation kind data of import system to be carried out pre-service with pretreatment module 3;

Step5: the data in single working procedure quality module 6 combined process standard databases 12 are carried out the evaluation of process capability index to each technology of pretreated cigarette working procedure;

Step6: the technique process quality of 7 pairs of cigarette working procedures of working procedure quality overall evaluation module of grey correlation analysis is carried out comprehensive evaluation;

Step7: evaluation result is showed to be connected with output module 8 and evaluation result is sent to output device 13 and is saved in the evaluation result database 14;

Among the described Step5 each technology of pretreated cigarette working procedure is carried out the evaluation of process capability index, may further comprise the steps:

(1): by the every technological parameter of instrument detecting cigarette, with technological parameter data recording, storage;

(2): reject the wrong or special sample in the technological parameter data;

(3): with the technological standards of every technological parameter is that standard center M and specification tolerance T store;

(4): technological parameter according to estimating the purpose grouping, for example will be compared the good and bad situation of different month process qualitys, and then the technological parameter in each month is as one group, and every group as an evaluation object;

(5): calculate the sample average x and the sample standard deviation σ of each each technological parameter of evaluation object respectively, computing formula is:

x＝∑x _i/n

$\mathrm{\&sigma;}=\sqrt{\mathrm{\&Sigma;}{(x_i-x)}^2/(n-1)}$

Wherein n is a number of samples in each evaluation object;

(6): with sample average x, sample variance σ, standard center M, upper specification limit T _U, specification lower limit T _LWith specification tolerance T substitution process capability index computing formula, calculate single process capability index X of each evaluation object _i(i=1,2, Λ, m), m is the number of evaluation object,

$C_{\mathrm{pk}}=\frac{|T_U-T_L|}{6\mathrm{\&sigma;}}(1-k)=\frac{\mathrm{<figure-callout\; id="6"\; label="T"\; filenames="A200810237698E00161.png"\; state="\{\{state\}\}">T</figure-callout>}}{6\mathrm{\&sigma;}}(1-k)$

Wherein $k=\frac{2(M-x)}{|T_U-T_L|}=\frac{2(M-x)}{T}$

Single process capability index X of each evaluation object _iThe row vector of forming by the Cpk of each technological parameter.

Technique process quality to cigarette working procedure among the described Step6 is carried out comprehensive evaluation, may further comprise the steps:

(1): with single process capability index X of each evaluation object _i(i=1,2, Λ, m) as with reference to ordered series of numbers, X _i(i=1,2, Λ, m) the capable vector formed of the maximal value of every row ordered series of numbers Y as a comparison, wherein m is the number of evaluation object;

(2): to reference sequence X _i(i=1,2, Λ m) carries out nondimensionalization with comparison ordered series of numbers Y with the initial value method and handles, and the ordered series of numbers after the processing is designated as respectively

With Y ＇

$X_i^{\&prime;}\left(k\right)=X_i\left(k\right)/Y\left(k\right)$

Y＇(k)＝Y(k)/Y(k)

Wherein k represents k technological parameter;

(3): ask

Each element and Y ＇ when the difference in prostatitis ${\mathrm{\&Delta;}}_i\left(k\right)=|X_i^{\&prime;}\left(k\right)-Y\&prime;\left(k\right)|$ (i=1,2, Λ, m), Δ _i(k) form matrix of differences;

(4): obtain the minimum value a and the maximal value b of matrix of differences, make resolution ratio ρ=0.5, compute associations coefficient r _i(k):

$r_i\left(k\right)=\frac{a+\mathrm{\&rho;b}}{{\mathrm{\&Delta;}}_i\left(k\right)+\mathrm{\&rho;b}}$

R wherein _i(k) expression

Degree of correlation with Y ＇ (k);

(5): ask grey relational grade γ _i(i=1,2, Λ, m).Obtain the integrated value of each evaluation object correlation coefficient:

${\mathrm{\&gamma;}}_i=\frac{1}{n}\underset{k}{\mathrm{\&Sigma;}}{\mathrm{\&gamma;}}_i\left(k\right)\mathrm{\&omega;}$

Wherein ω is the weight of the technological parameter of evaluation object;

(6): the grey relational grade to each evaluation object sorts, and obtains the quality ordering of the process quality integrated value of each evaluation object.

In conjunction with the accompanying drawings 1 and accompanying drawing 2 be that the present invention will be described in detail for example with the comprehensive evaluation of cigarette technology for making tobacco threds:

1. User login

The user inputs user name, password, role, and the system searching user right satisfies condition, and then allows this user further to operate, otherwise forces this user to withdraw from native system.

2. data importing

For the cigarette technology for making tobacco threds is carried out comprehensive evaluation, need the related data of technology for making tobacco threds to import to system.Therefore the sample data with the cigarette technology for making tobacco threds of certain cigarette factory trade mark of instrument detecting imports in the system.Herein, also can add data, operation such as deletion and renewal.

3. the operation kind is selected

The operation kind that selection will be estimated from operation kind class template.Also need set evaluation method this moment: by estimating in month, estimate per year etc.This example is selected to estimate by month, and selects loosening and gaining moisture moisture, baking posterior lobe silk moisture, and the dry two wires of stem moisture, dry line moisture of stem and the mixed silk of perfuming moisture are 5 technological parameters.

4. data are selected and pre-service

Choose the sample data of the cigarette technology for making tobacco threds of this trade mark of this cigarette factory January, February, March, September And October, and data are carried out pre-service work such as singular value rejecting, missing values processing, reach the purpose of optimizing the input data.Partial data sees Table 1;

The part raw data of certain cigarette factory trade mark silk producing procedures of table 1

Throwing batch	Moisture behind the loosening and gaining moisture	5k baking posterior lobe silk moisture	Two wires moisture after the stem drying	A line moisture after the stem drying	Perfuming silk union moisture
						0707BG####001	20.426	12.591	13.804	13.832	12.255
0707BG####002	20.153	12.584	13.804	13.832	12.345
						0707BG####003	20.344	12.596	13.858	13.831	11.9
0707BG####00	20.054	12.592	13.858	13.831	11.94

4
						0707BG####005	19.647	12.625	13.858	13.831	12.02
… …	… …	… …	… …	… …	… …
						0803BG####004	19.19	12.897	13.959	13.992	12.645
0803BG####005	19.124	12.886	13.947	13.964	12.535
						0803BG####006	18.954	12.881	13.876	13.955	12.315
0803BG####007	18.926	12.897	13.876	13.955	12.385
						0803BG####008	18.892	12.869	13.876	13.955	12.565

System reads this more than trade mark automatically from the technological standards database 5 months the technological standards data of each technological parameter.See Table 2;

The standard of certain cigarette factory trade mark silk producing procedures of table 2

5. single process capability analysis phase

Each technology of cigarette working procedure is carried out the evaluation of process capability index with single working procedure quality module.This is one of gordian technique of native system, has finished following algorithm in internal system:

5.1 because this example selects is to estimate by month, system is according to the throwing batch information, with the sample data of this trade mark of this cigarette factory January, February, March, September And October respectively with X _i(i=1, January; I=2, March; I=3, March; I=4, September; I=5, October) be expressed as five evaluation objects.Loosening and gaining moisture moisture, baking posterior lobe silk moisture, the dry two wires of stem moisture, dry line moisture of stem and the mixed silk of perfuming moisture are evaluation index for 5 technological parameters (being operation) that influence the cigarette technology for making tobacco threds.

5.3, obtain its standard center vector M and specification tolerance vector T according to the technology for making tobacco threds standard of table 2;

5.4 the sample data of five evaluation objects as input value, is calculated their sample average x and sample standard deviation σ respectively;

5.5, calculate single process capability index X of each evaluation object with sample average x, sample standard deviation σ, standard center M and specification tolerance T substitution process capability index computing formula.The results are shown in Table 3;

The process capability index of certain the cigarette factory trade mark each parameter of five months of table 3

6. grey correlation analysis comprehensive evaluation stage

Working procedure quality overall evaluation module with grey correlation analysis is carried out comprehensive evaluation to the technique process quality of cigarette working procedure.This be native system gordian technique two, finished following algorithm in internal system:

6.1 single process capability index of table 3 as the reference ordered series of numbers, is represented { X with X _i}={ X ₁, X ₂, X ₃, X ₄, X ₅, the capable vector that the maximal value of the every row of X is formed is ordered series of numbers as a comparison, represents { Y with Y _i}={ Y ₁, Y ₂, Y ₃, Y ₄, Y ₅.

Handle 6.2 reference sequence X is carried out nondimensionalization with comparison ordered series of numbers Y with the initial value method, obtain new reference sequence X ＇ and Y ＇;

Ask absolute difference, Δ 6.3 ask each element of X ＇ and the prostatitis of working as of Y ＇ _i(k)=| X _i(k)-Y ＇ (k) |;

6.4 obtain the matrix of differences Δ _i(k) minimum value a=0 and maximal value b=0.6422 make resolution ratio ρ=0.5, ask the degree of correlation r that obtains each evaluation object and each technological parameter _i(k), see Table 4;

6.5 try to achieve grey relational grade γ _iThereby, obtaining the integrated value of each evaluation object correlation coefficient, the value of ω is determined by the user, for example can be made as ω=[0.5,1.5,0.5,0.5,2];

6.6 the grey relational grade to each evaluation object sorts at last, the quality ordering that obtains the working procedure quality integrated value of each evaluation object sees Table 4.

The correlation coefficient of each evaluation object of table 4, grey relational grade and ranking results

7. evaluation result is showed and output module

Form with form and figure is showed the evaluation result of table 3 and table 4, makes the easier interpretation of result of carrying out next step of user.The chart that the user shows by evaluation result, more clear and systematically evaluation result is analyzed.

The user can be saved in the chart of current evaluation result and user's oneself analysis result in the database, convenient inquiry, application and further analysis from now on.

To the result of this cigarette factory this technology for making tobacco threds comprehensive evaluation of five months of this trade mark cigarette, the order of quality that can determine five evaluation objects (five months) is X ₂X ₅X ₁X ₃X ₄, we can analyze comparison according to the result of this good and bad ordering and in conjunction with the process quality comprehensive condition of each evaluation object, thereby instruct production practices with the accessible conclusion of user.As X ₂The process quality comprehensive condition of (February) technology for making tobacco threds is best, meets the standard-required of explained hereafter most; X ₄The process quality comprehensive condition of (September) technology for making tobacco threds is the poorest in five evaluation objects, and some factor in this process of manufacture is described in month, and for example production equipment, personnel operation, raw material, production environment etc. have much room for improvement.

Claims (10)

1. based on the cigarette working procedure quality overall evaluation system of grey correlation analysis, it is characterized in that: comprise an authority management module (1), described authority management module (1) is connected with data importing module (2); Described data importing module (2) is selected to be connected with pretreatment module (3) with data, and data are selected with pretreatment module (3) data to be carried out pre-service; Described data are selected to be connected with single working procedure quality module (6) with pretreatment module (3); Described single working procedure quality module (6) is connected with the working procedure quality overall evaluation module (7) of grey correlation analysis; The working procedure quality overall evaluation module (7) of described grey correlation analysis is showed with evaluation result and is connected with output module (8); Described data are selected also to select module (4) with the operation kind with pretreatment module (3); Described single working procedure quality module (6) also is connected with technological standards database (12).

2. the cigarette working procedure quality overall evaluation system based on grey correlation analysis according to claim 1, it is characterized in that: described authority management module (1) is connected with rights database (9) and carries out data transfer, and the role in the system, user, user's group are defined and the management of authority separately.

3. the cigarette working procedure quality overall evaluation system based on grey correlation analysis according to claim 1, it is characterized in that: described data importing module (2) is connected with technological parameter database (10) carries out data transfer, in the data importing system, and can add data, operation such as deletion, renewal, the real-time management data.

4. the cigarette working procedure quality overall evaluation system based on grey correlation analysis according to claim 1, it is characterized in that: described operation kind selection module (4) is connected with operation kind template base (11) carries out data transfer, select corresponding working procedures from operation kind template base (11), and can create new operation kind.

5. the cigarette working procedure quality overall evaluation system based on grey correlation analysis according to claim 1 is characterized in that: methods such as operation kind that described data are selected and pretreatment module (3) basis is selected or singular value, repetition values, missing values are carried out pre-service to data.

6. the cigarette working procedure quality overall evaluation system based on grey correlation analysis according to claim 1 is characterized in that: described technological standards database (12) is connected with technological standards administration module (5) and carries out data transfer; Described technological standards administration module (5) provides preservation, operations such as modification.

7. the cigarette working procedure quality overall evaluation system based on grey correlation analysis according to claim 1, it is characterized in that: described evaluation result is showed and output module (8) passes to output device (13) with the result, intuitively shows result with the form of chart; Output device (13) passes to evaluation result database (14) again with data.

8. cigarette working procedure quality overall evaluation method based on grey correlation analysis, it is characterized in that: this method may further comprise the steps:

Step1: the user inputs user name, password, role, and authority management module (1) is according to the data search user right in the rights database (9).Satisfy condition, then allow this user further to operate, otherwise force this user to withdraw from native system;

Step2: the user by data importing module (2) with in the data importing system in the technological parameter database (10);

Step3: the operation kind that the user selects module (4) will estimate from operation kind template base (11) selection by the operation kind, and send the data to data selection and pretreatment module (3);

Step4: data are selected according to the operation kind data of import system to be carried out pre-service with pretreatment module (3);

Step5: the data in single working procedure quality module (6) combined process standard database (12) are carried out the evaluation of process capability index to each technology of pretreated cigarette working procedure;

Step6: the working procedure quality overall evaluation module (7) of grey correlation analysis is carried out comprehensive evaluation to the technique process quality of cigarette working procedure;

Step7: evaluation result is showed to be connected with output module (8) and evaluation result is sent to output device (13) and is saved in the evaluation result database (14).

9. a kind of cigarette working procedure quality overall evaluation method based on grey correlation analysis according to claim 7 is characterized in that:

Among the described Step5 each technology of pretreated cigarette working procedure is carried out the evaluation of process capability index, may further comprise the steps:

(1): by the every technological parameter of instrument detecting cigarette, with technological parameter data recording, storage;

(2): reject the wrong or special sample in the technological parameter data;

(3): with the technological standards of every technological parameter is that standard center M and specification tolerance T store;

(4): technological parameter according to estimating the purpose grouping, for example will be compared the good and bad situation of different month process qualitys, and then the technological parameter in each month is as one group, and every group as an evaluation object;

(5): calculate the sample average x and the sample standard deviation σ of each each technological parameter of evaluation object respectively, computing formula is:

x＝∑x _i/n

$\mathrm{\&sigma;}=\sqrt{\mathrm{\&Sigma;}{(x_i-x)}^2/(n-1)}$

Wherein n is a number of samples in each evaluation object;

(6): with sample average x, sample variance σ, standard center M, upper specification limit T _U, specification lower limit T _LWith specification tolerance T substitution process capability index computing formula, calculate single process capability index X of each evaluation object _i(i=1,2, Λ, m), m is the number of evaluation object,

$C_{\mathrm{pk}}=\frac{|T_U-T_L|}{6\mathrm{\&sigma;}}(1-k)=\frac{T}{6\mathrm{\&sigma;}}(1-k)$

Wherein $k=\frac{2(M-x)}{|T_U-T_L|}=\frac{2(M-x)}{T}$

Single process capability index X of each evaluation object _iThe row vector of forming by the Cpk of each technological parameter.

10. a kind of cigarette working procedure quality overall evaluation method based on grey correlation analysis according to claim 7 is characterized in that:

Technique process quality to cigarette working procedure among the described Step6 is carried out comprehensive evaluation, may further comprise the steps:

(1): with single process capability index X of each evaluation object _i(i=1,2, Λ, m) as with reference to ordered series of numbers, X _i(i=1,2, Λ, m) the capable vector formed of the maximal value of every row ordered series of numbers Y as a comparison, wherein m is the number of evaluation object;

(2): to reference sequence X _i(i=1,2, Λ m) carries out nondimensionalization with comparison ordered series of numbers Y with the initial value method and handles, and the ordered series of numbers after the processing is designated as respectively

And Y

$X_i^{\&prime;}\left(k\right)=X_i\left(k\right)/Y\left(k\right)$

Y′(k)＝Y(k)/Y(k)

Wherein k represents k technological parameter;

(3): ask

Each element and Y ' when the difference in prostatitis ${\mathrm{\&Delta;}}_i\left(k\right)=|X_i^{\&prime;}\left(k\right)-Y\&prime;\left(k\right)|$ (i=1,2, Λ, m), Δ _i(k) form matrix of differences;

(4): obtain the minimum value a and the maximal value b of matrix of differences, make resolution ratio ρ=0.5, compute associations coefficient r _i(k):

$r_i\left(k\right)=\frac{a+\mathrm{\&rho;b}}{{\mathrm{\&Delta;}}_i\left(k\right)+\mathrm{\&rho;b}}$

R wherein _i(k) expression

And Y ' degree of correlation (k);

(5): ask grey relational grade γ _i(i=1,2, Λ, m).Obtain the integrated value of each evaluation object correlation coefficient:

${\mathrm{\&gamma;}}_i=\frac{1}{n}\underset{k}{\mathrm{\&Sigma;}}{\mathrm{\&gamma;}}_i\left(k\right)\mathrm{\&omega;}$

Wherein ω is the weight of the technological parameter of evaluation object;

(6): the grey relational grade to each evaluation object sorts, and obtains the quality ordering of the process quality integrated value of each evaluation object.

Images