CN110033201A - A kind of tobacco industry batch overall process quality testing and improved method and apparatus - Google Patents
A kind of tobacco industry batch overall process quality testing and improved method and apparatus Download PDFInfo
- Publication number
- CN110033201A CN110033201A CN201910322172.3A CN201910322172A CN110033201A CN 110033201 A CN110033201 A CN 110033201A CN 201910322172 A CN201910322172 A CN 201910322172A CN 110033201 A CN110033201 A CN 110033201A
- Authority
- CN
- China
- Prior art keywords
- data
- evaluation index
- index
- class
- tobacco industry
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The invention discloses a kind of tobacco industry batch overall process quality testing and improved method and apparatus, comprising: data acquisition phase: acquisition production environment data, production process data, production equipment technological parameter data, production process quality detecting data；The data quality accessment stage: every class data are directed to, calculate at least one data evaluation index corresponding with every class data, wherein data evaluation index includes integrality, accuracy, consistency, timeliness, the 7 class data evaluation index of uniqueness, validity and stability of data, and the overall target of acquisition data is calculated according to 7 class data evaluation indexs；Quality of data diagnosis and improvement stage: when aggregation of data index is unsatisfactory for healthiness condition, the data evaluation index of problem is then reversely deduced according to aggregation of data index, and improves mechanism according to data and amendment is improved to that the corresponding data of class data evaluation index to go wrong.This method and device can be realized evaluation, diagnosis and amendment to lot data.
Description
Technical field
The invention belongs to tobacco industry batch administrative skill fields, particularly belong to a kind of tobacco industry batch overall process data
Quality evaluation and improved method and apparatus.
Background technique
With IT application in enterprises, digitlization, intelligentized deep development, data have become new production factors, and
Very important effect is embodied in enterprise.The confidence level and availability for how improving data, play the value of data assets, are
Realize the critical issue of traditional industries digitlization transition.
So-called batch management system is both the management system of batch during being directed to material circulation.For batch, because
The difference of demand also produces different way to manages so having different definition.On the whole, batch management system is
Enterprise realizes the important system of product lifecycle retrospect, multi dimensional analysis and fine-grained management, is the circulation process of material
In indispensable system.The quality of data is the key foundation of batch management.
For a long time, since business tie-up produces the factors such as span is big, operation flow is complicated, data volume is big, batch management
Often there is the problems such as shortage of data is imperfect, data are inconsistent, data feedback lags not in time in system.Most enterprises are to itself
The verification index mode based on data different characteristics is generallyd use with co-production enterprise, from data integrity, timeliness, accurate
Property, the dimensional properties such as consistency define multiple quantizating index and carry out data monitorings, but data monitoring system is not perfect, numerous fingers
Mark is difficult to macroscopically reflect the quality condition of entire data, it is difficult to carry out comprehensive assessment to the quality of data, can not provide auxiliary and determine
Plan analysis, and lack corresponding diagnosis and improve mechanism.
The patent application of 101894319 A of application publication number CN discloses a kind of tobacco enterprise data quality management system
And method comprising: information acquisition module, it is various original required for quality rule management and data quality diagnosis for collecting
Information；Quality rule management module, for managing all objects in accuracy, consistency, integrality, timeliness, accessibility
The quality rule that should be followed in terms of this five big quality metric；Quality of data diagnostic module, for being passed according to information acquisition module
The quality rule definition and specific quality of data diagnostic task, timing of the data, quality rule management module passed carry out data
Quality diagnosis simultaneously generates quality diagnosis result；Quality of data reporting modules, the result information for diagnosing the quality of data is not with
The user is passed to the mode that user uses；Quality of data processing module, for according to the quality of data reporting modules
Listed quality problems inventory to be processed, according to the high reduction process quality problems of rank.Although the data quality management system is open
By for statistical analysis to acquisition data, to judge the quality of data, but it is many to the means of the statistical analysis of data,
The index of statistics is also not quite similar and the specific characteristic manner of index is different, causes the quality of data finally evaluated also can be thousand poor
Ten thousand are not, therefore it is not clear whether be able to achieve the accurate evaluation to the quality of data in data quality management system, in other words, the number
According to quality control system, there is no open to accurate review number according to the technological means of quality.
Summary of the invention
Of the invention is to provide a kind of tobacco industry batch overall process quality testing and improved method and apparatus, should
Method and apparatus can integrate the integralities of tobacco industry batch overall process data, accuracy, consistency, timeliness, uniqueness,
Validity and stability fully assess the quality of data, and the amendment of data is acquired according to the quality of data.
For achieving the above object, the present invention the following technical schemes are provided:
On the one hand, a kind of tobacco industry batch overall process quality testing and improved method, including following procedure:
Data acquisition phase: acquisition production environment data, production process data, production equipment technological parameter data, production
Process quality detecting data；
The data quality accessment stage: being directed to every class data, calculates at least one data evaluation corresponding with every class data and refers to
Mark, wherein data evaluation index includes the integralities of data, accuracy, consistency, timeliness, uniqueness, validity and steady
Qualitative 7 class data evaluation index, and the overall target for acquiring data is calculated according to 7 class data evaluation indexs；
Quality of data diagnosis and improvement stage: when aggregation of data index is unsatisfactory for healthiness condition, then according to aggregation of data
Index reversely deduces the data evaluation index of problem, and improves mechanism according to data to that class data evaluation index to go wrong
Corresponding data improve amendment, wherein data improve mechanism include: missing data amended record and passback, repeated data
Cleaning and the warning not in time of check, the rejecting of invalid data, the amendment of logic error, the processing of inconsistent data, data, number
According to the limitation of wide fluctuations.
On the other hand, a kind of tobacco industry batch overall process quality testing and improved device, which is characterized in that packet
It includes:
Data acquisition module: for acquire production environment data, production process data, production equipment technological parameter data,
Production process quality detecting data；
Data quality accessment module: being directed to every class data, calculates at least one data evaluation corresponding with every class data and refers to
Mark, wherein data evaluation index includes the integralities of data, accuracy, consistency, timeliness, uniqueness, validity and steady
Qualitative 7 class data evaluation index, and the overall target for acquiring data is calculated according to 7 class data evaluation indexs；
Quality of data diagnosis and improve module: for when aggregation of data index is unsatisfactory for healthiness condition, then according to data
Overall target reversely deduces the data evaluation index of problem, and improves mechanism according to data to that class data evaluation to go wrong
The corresponding data of index improve amendment, wherein data improve mechanism include: missing data amended record and passback, repeat number
According to cleaning and check, the rejecting of invalid data, the amendment of logic error, the processing of inconsistent data, the police of data not in time
It accuses, the limitation of data wide fluctuations.
The present invention realizes the pipe that the processes such as the acquisition of lot data, assessment, diagnosis and improvement are realized with the quality of data
Reason, specifically, the quality of data can be determined according to 7 evaluation indexes of data and the overall target of data, improves meter
The accuracy for calculating the quality of data, while judging the quality of data according to the healthiness condition of setting, to unsound data into
Row amendment, ensure that the quality of management industrial data.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to do simply to introduce, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art, can be with root under the premise of not making the creative labor
Other accompanying drawings are obtained according to these attached drawings.
Fig. 1 is batch overall process quality testing and the procedure chart that improved method and apparatus are realized；
Fig. 2 is batch management quality testing system schematic diagram；
Fig. 3 is that the batch management quality of data improves module diagram.
Specific embodiment
To make the objectives, technical solutions, and advantages of the present invention more comprehensible, with reference to the accompanying drawings and embodiments to this
Invention is described in further detail.It should be appreciated that the specific embodiments described herein are only used to explain the present invention,
And the scope of protection of the present invention is not limited.
A kind of tobacco industry batch overall process quality testing and improved method and apparatus are present embodiments provided, it should
Method and apparatus can be applied in tobacco leaf production.
As shown in Figure 1, tobacco industry batch overall process quality testing provided in this embodiment and improved method packet
Include following procedure:
Data acquisition phase, it is main acquire production environment data, production process data, production equipment technological parameter data,
Production process quality detecting data.
Under normal circumstances, mainly the acquisitions production such as production equipment, barcode scanning equipment, quality inspection equipment, environment measuring equipment
Various types of data in the process.
During tobacco leaf production, it is related to:
Material supply link mainly goes out to be put in storage the equipment such as barcode scanning equipment, cored temperature sensor acquisition object by goods and materials
The data in supply stage are provided, such as goods and materials go out to be put in storage lot data, piece cigarette cored temperature, insect pest situation；
Production inventory link mainly obtains storage facility located at processing plant with sweeping decoding apparatus, the fixed barcode scanning terminal of shelf magazine by auxiliary material
In stockpile number, shelf magazine enter and leave lot data；
Leaf processed feeds intake link, mainly passes through the phase that the equipment such as leaf production line processed and fixed barcode scanning terminal obtain the leaf stage processed
Close data, as piece cigarette feed intake lot data, enter and leave blade cabinet lot data, vacuum conditioning moisture, charge temperature, casing drum turn
Speed, cut before leaves moisting flow etc.；
Fiber tow production link mainly obtains the dependency number in fiber tow production stage by scrap prodn. line and fixed barcode scanning terminal
According to, such as cut tobacco vanning, mould turnover and pipe tobacco vanning, mould turnover lot data, discrepancy cut tobacco, pipe tobacco shelf magazine lot data, steam
Pressure, vapour volume flow, cut tobacco degree of purity, hot blast temperature etc.；
Filter stick production link mainly obtains filter stick life by filter stick formation board, filter stick transmitter, filter stick RFID terminal
The related data in production stage, as filter stick production quantity, weight, filter stick dress lattice enter and leave filter stick shelf magazine lot data, filter stick transmitting
To cigarette machine lot data, roller speed ratio, filter stick hardness, filter-rod periphery etc..
Volume wraps the link that feeds intake, and mainly passes through the correlation of the acquisitions such as mica wrappingmachine platform and hand-held barcode scanning equipment volume packet production phase
Data, such as the batch of auxiliary material of cigarette paper, cork paper, barrel, trade mark minimum package unit feed intake data, hungry, gas leakage, light
Hold rejecting, soldering iron actual temperature, host speed, subsidiary engine speed, embossing roller pressure etc.；
Finished product allots link, mainly obtains finished product by holding barcode scanning equipment, fixed barcode scanning equipment and RFID terminal
Data are alloted, such as part cigarette and pallet binding data, the finished parts cigarette disk that puts one's child in a boarding nursery enters and leaves finished product shelf magazine lot data, enters and leaves finished bin
Library lot data etc.；
In addition, it is additionally provided with some temperature sensors, humidity sensor etc., and for acquiring production environment data, such as each monitoring
The temperature of point, humidity data.It is acquired and is produced by quality inspections equipment such as gas chromatograph, smoking machine, QTM, head space gas chromatograph-mass spectrometers
Process quality detecting data, such as cigarette finished product circumference, weight, resistance to suction, hardness and total ventilation rate.Flavors and fragrances relative density, fusing point
With acid value etc..Material benzene content, air permeability and tensile strength etc..Raw material nicotine, total reducing sugar and thickness etc..
The data quality accessment stage: being directed to every class data, calculates at least one data evaluation corresponding with every class data and refers to
Mark, wherein data evaluation index includes the integralities of data, accuracy, consistency, timeliness, uniqueness, validity and steady
Qualitative 7 class data evaluation index, and the overall target for acquiring data is calculated according to 7 class data evaluation indexs.
According to actual industrial scene, the data distribution characteristic and attribute of every class data be not identical, and assessment data is caused to refer to
Mark can also be not quite similar, and therefore, when calculating data evaluation index, according to the characteristic of data, corresponding evaluation index be selected
Evaluate such data.
By a large amount of practical summary, integrality, accuracy, consistency, timeliness, uniqueness, validity and stabilization
7 class data evaluation indexs of property can comprehensively evaluate the quality of data.Specifically,
The integrality refers to whether goods and materials, production, the data of three aspects of finished product are complete, and whether unit dimension is complete；
From the business perspective, it is mainly measured in terms of the different grain size of batch management, including record of production content intact, unit
Dimension is complete, and operation quantity is complete etc., therefore defining must have rate as the index for measuring integrality, must there is the formula of rate index
It is as follows:
Specifically, integrity metrics are defined as follows:
Wherein, W is integrity metrics, and n is the sum of material, miFor the practical injected volume of i-th of material, m0iIt is i-th
The theoretical injected volume of material, ωiIt is adaptive weighting for i-th of material, and
Wherein, the calculating process of adaptive weighting are as follows:
(a) the weight distribution of each each material of the trade mark is initialized, each material most starts to be endowed identical weight, i.e.,Initial weight distribution
(b) iteration j=1 ... m, statistics material are unsatisfactory for the number e of desired value in the single trade markj,
Wherein, E (MHR1i) be material expectation must have rate MHR；
(c) the calculated result factor alpha of the trade mark is calculatedj, the more big then MHR of coefficient is higher；
(d) it updates weight and is distributed Dj+1；
When material is unsatisfactory for expectation MHR,
When material meets expectation MHR,
Wherein, ZjFor normaliztion constant
The accuracy refers to whether data are accurate in the service logic in terms of each dimension, frequency, product back-tracing
Whether traceability chain penetrates through, therefore defines retrospect error rate as the index for measuring accuracy, specifically, accuracy is defined as:
Wherein, P is accuracy index.
The consistency, which refers to, to be uploaded to the data of batch management system (ERP guarantees number by periodic review with ERP system
According to accurate) data it is whether consistent, in production of cigarettes, the especially consistency of inventory information, therefore define batch inventory
Concordance rate is as the index for measuring consistency, specifically, conformance definition are as follows:
Wherein, C is coincident indicator.
Generally, the value of data has regular hour property, and the timeliness refers to whether data are full in business procedure
Whether sufficient business reports according to the renewal frequency of regulation the timeliness demand and related data of data；Specifically, and
Shi Xing is defined as:
Wherein, S is timeliness, tdFor the lag time that timeliness constraint rule defines,For average report cycle, formula
Are as follows:
Wherein, tinFor data inputting time, toccurFor data time of origin, NrecordFor logging data amount, p is data record
The number entered, q are the number of data set.
The uniqueness refers in across enterprise batch management process that data cannot repeat to upload and be recorded, and needs to guarantee
The unique of data is uploaded, in production of cigarettes, the uniqueness of batch number is extremely important, and puts into the quantity and output quantity of material
There should be certain corresponding relationship, therefore define Data duplication rate as the index for measuring uniqueness；Specifically, batch input and output
Data duplication rate, that is, unique definition are as follows:
Wherein, R1For uniqueness, NallFor all Mission Numbers, NrepeatFor duplicate Mission Number.
Wherein, PinFor input quantity, PoutFor output quantity, η is the input output theory ratio system of manufacturing technique requirent
Number.
The validity refers to that data are effective in threshold range, and beyond the data after threshold value be it is invalid or
Mistake；General batch processes data validity index is measured, specifically, validity is defined as:
Wherein, V is validity, and the data that x (i) is i-th, S is the total amount of data,xWithThe respectively threshold of data area
It is worth lower and upper limit.
The stability refers in the production process of enterprise, reflects fluctuation situation of the process data in threshold range,
Therefore batch processes stability is defined as the index for measuring stability；Specifically, definition of stability are as follows:
Wherein,xWithThe respectively bottom threshold and the upper limit of data area.
The validity refers to that data are effective in some threshold range, and is invalid beyond the data after threshold value
Or mistake.Therefore data effective percentage is defined as the index for measuring validity.
Comprehensive estimation method carries out overall merit to cooperation manufacturing enterprise's batch management quality of data, and macroscopically reflection is each
The aggregate level of the batch Management System Data quality of a co-production enterprise.On the basis of obtaining 7 class data evaluation index,
The overall target for calculating data comes the quality of overall merit data, specifically, the calculating process of the overall target are as follows:
(a) integrality, accuracy, consistency, timeliness, uniqueness, validity and 7 class data evaluation of stability are referred to
Mark carries out forward directionization processing；
Since different indexs is divided into large and minimal type, such as must there are rate, concordance rate to be the bigger the better, and repetitive rate,
Error rate is then the smaller the better, it is therefore desirable to carry out unification processing to index.For negative sense index, converted using following formula
For positive type:
(b) it is compared scale using significance level of the method for expert estimation to data evaluation index, building one is sentenced
Disconnected matrix A (aij)n×n, judgment matrix A (aij)n×nIn, the importance value of each element representation data evaluation index, and each member
Element is about diagonal line at reciprocal relation；
(c) judge judgment matrix A (aij)n×nMaximum eigenvalue λmaxCorresponding feature vector, then feature vector is carried out
Normalized, obtained vector are the subjective weight vectors w=(ω of each data evaluation index1,...,ωn)；
Due to judgment matrix A (aij)n×nOrder it is generally higher, it is more difficult directly to calculate its characteristic value, therefore this hair
It is bright that weight, expression formula are calculated using geometric average method are as follows:
The maximum eigenvalue of judgment matrix can be by weight vectors approximate calculation:
Wherein, A is judgment matrix, and w is weight vectors, ωiFor i-th of weight vectors, n is the quantity of evaluation index.
By carry out consistency check determine the mutual important ratio of index compared with logical consistency, be that step analysis conclusion can
By whether premise.The inconsistent degree CI of matrix can be calculated according to obtained maximum eigenvalue:
Finally obtain random consistency ratio CR
Wherein, RI is the random index of matrix, and value is as shown in table 1 below:
The corresponding RI value of 2 different rank of table
As the random consistency ratio CR < 0.10 of judgment matrix, then illustrate that there is concordance rate can satisfy and want for matrix
It asks.
(d) the entropy weight vector for calculating every class data evaluation index, specifically, for individual data evaluation index, information
Entropy are as follows:
Wherein,M is all quantity for participating in evaluation, yijI-th of data evaluation for j-th of unit refers to
Mark；
The entropy weight weight ε of i-th of indexi:
(e) it is directed to each data evaluation index, the entropy weight that the subjective weight and step (d) that step (c) is obtained obtain is weighed
It is combined again, obtains the comprehensive weight of each data evaluation index；
Specifically, subjective weights omega analytic hierarchy process (AHP) obtainediThe entropy weight weight ε obtained with entropy assessmentiCarrying out synthesis can
To obtain the synthetic weights weight values α of each indexi, due to αiValue should be with subjective weights omegaiWith entropy weight weight εiValue it is close as far as possible,
Meet following formula:
Solving the above extreme-value problem using Lagrange multiplier method can obtain:
(f) the comprehensive evaluation index F of data is calculated according to the comprehensive weight of data evaluation indexj:
Wherein, yijFor i-th of data evaluation index of j-th of unit, αiFor the synthetic weights of i-th of data evaluation index
Weight.
Quality of data diagnosis and improvement stage: when aggregation of data index is unsatisfactory for healthiness condition, then according to aggregation of data
Index reversely deduces the data evaluation index of problem, and improves mechanism according to data to that class data evaluation index to go wrong
Corresponding data improve amendment, wherein data improve mechanism include: missing data amended record and passback, repeated data
Cleaning and the warning not in time of check, the rejecting of invalid data, the amendment of logic error, the processing of inconsistent data, data, number
According to the limitation of wide fluctuations.
Healthiness condition can refer to the overall target threshold value of setting, work as satisfactionWhen, i.e. overall target FjLower than some
Overall target threshold valueWhen, i.e. aggregation of data index is unsatisfactory for healthiness condition, needs to improve data, i.e., reverse-direction derivation has
The data evaluation index of problem, specifically, the data evaluation index packet that problem is reversely deduced according to aggregation of data index
It includes:
Every class data evaluation index is obtained according to the calculating process reverse search of overall target；
For every class data evaluation index, using the threshold ratings of such data evaluation index, such data evaluation index is
It is no to go wrong.
During production of cigarettes, as shown in figure 3, improved process are as follows: the quality of data is improved by the first computing unit, industry
Business system unit, edge calculations unit, apparatus control system, batch application server, lot data library server, data exchange
The composition such as bus, second computing unit.First computing unit is directed to the business datum triggering error correction for being unsatisfactory for overall target requirement
The single index to go wrong in 7 index dimensions is inversely found in instruction, descending due to each single index weighted
Successively judge whether corresponding single index goes wrong；Operation system unit is classified according to operation system, finds out data appearance
The type of service of problem, including but not limited to: volume packet feeds intake, leaf processed feeds intake, fiber tow production, production inventory, finished product are alloted；Side
Edge computing unit is traced by the equipment that type of service reflects problem-indicator, finds problem source, generates issue list；
Corresponding measure is triggered according to issue list, extracts quality improvement mechanism from the data rule library of lot data library server automatically
And method, including but not limited to:
A) when integrity metrics are unsatisfactory for requiring, by inversely tracing positioning, relevant device carries out the benefit of missing data
Record and passback
B) when accuracy index is unsatisfactory for requiring, the amendment of logic error is carried out
C) when uniqueness index is unsatisfactory for requiring, the cleaning and check of repeated data are carried out
D) when Validity Index is unsatisfactory for requiring, the rejecting of invalid data is carried out
E) when coincident indicator is unsatisfactory for requiring, the processing of inconsistent data is carried out
F) when timeliness index is unsatisfactory for requiring, the warning of data not in time is issued to equipment
G) when stability indicator is unsatisfactory for requiring, the limitation of data wide fluctuations and the prediction of data variation are carried out
Deng
The reasonable computation of Scientific Establishment and index weight value by quality of data index, batch overall process data of the invention
Batch manages the quality of data between quality evaluation and improved method more accurately can objectively measure different co-production enterprises
Aggregate level.
By dignosis and improvement mechanism, batch overall process quality testing of the invention and improve device can automatically from
Extraction improves mechanism in data rule library and method, improvement and optimization lot data quality, raising enterprise's fine-grained management are horizontal.
The present embodiment additionally provides a kind of tobacco industry batch overall process quality testing and improved device, comprising:
Data acquisition module: for acquire production environment data, production process data, production equipment technological parameter data,
Production process quality detecting data；
Data quality accessment module: being directed to every class data, calculates at least one data evaluation corresponding with every class data and refers to
Mark, wherein data evaluation index includes the integralities of data, accuracy, consistency, timeliness, uniqueness, validity and steady
Qualitative 7 class data evaluation index, and the overall target for acquiring data is calculated according to 7 class data evaluation indexs；
Quality of data diagnosis and improve module: for when aggregation of data index is unsatisfactory for healthiness condition, then according to data
Overall target reversely deduces the data evaluation index of problem, and improves mechanism according to data to that class data evaluation to go wrong
The corresponding data of index improve amendment, wherein data improve mechanism include: missing data amended record and passback, repeat number
According to cleaning and check, the rejecting of invalid data, the amendment of logic error, the processing of inconsistent data, the police of data not in time
It accuses, the limitation of data wide fluctuations.
Tobacco industry batch overall process quality testing and improved device and above-mentioned tobacco industry batch overall process number
The function of realizing according to quality evaluation and improved method is identical with means and bring technical effect, and details are not described herein.
Technical solution of the present invention and beneficial effect is described in detail in above-described specific embodiment, Ying Li
Solution is not intended to restrict the invention the foregoing is merely presently most preferred embodiment of the invention, all in principle model of the invention
Interior done any modification, supplementary, and equivalent replacement etc. are enclosed, should all be included in the protection scope of the present invention.
Claims (10)
1. a kind of tobacco industry batch overall process quality testing and improved method, including following procedure:
Data acquisition phase: acquisition production environment data, production process data, production equipment technological parameter data, production process
Quality detecting data；
The data quality accessment stage: being directed to every class data, calculates at least one data evaluation index corresponding with every class data,
Middle data evaluation index includes integrality, accuracy, consistency, timeliness, uniqueness, validity and the stability 7 of data
Class data evaluation index, and the overall target for acquiring data is calculated according to 7 class data evaluation indexs；
Quality of data diagnosis and improvement stage: when aggregation of data index is unsatisfactory for healthiness condition, then according to aggregation of data index
The data evaluation index of problem is reversely deduced, and improves mechanism according to data and that class data evaluation index to go wrong is corresponded to
Data improve amendment, wherein data improve mechanism include: missing data amended record and passback, the cleaning of repeated data
It is big with the warning not in time of check, the rejecting of invalid data, the amendment of logic error, the processing of inconsistent data, data, data
The limitation of range fluctuation.
2. tobacco industry batch overall process quality testing as described in claim 1 and improved method, which is characterized in that
The integrality refers to whether goods and materials, production, the data of three aspects of finished product are complete, and whether unit dimension is complete；Specifically, complete
Whole property index definition is as follows:
Wherein, W is integrity metrics, and n is the sum of material, miFor the practical injected volume of i-th of material, m0iFor i-th of material
Theoretical injected volume, ωiIt is adaptive weighting for i-th of material, and
3. tobacco industry batch overall process quality testing as described in claim 1 and improved method, which is characterized in that
The accuracy refers to whether data are accurate in the service logic in terms of each dimension, frequency, and the traceability chain of product back-tracing is
No perforation, specifically, accuracy is defined as:
Wherein, P is accuracy index.
4. tobacco industry batch overall process quality testing as described in claim 1 and improved method, which is characterized in that
The consistency refers to whether the data of the data and ERP system that upload to batch management system are consistent, and specifically, consistency is fixed
Justice are as follows:
Wherein, C is coincident indicator.
5. tobacco industry batch overall process quality testing as described in claim 1 and improved method, which is characterized in that
The timeliness refer to data whether meet in business procedure business to the timeliness demand and related data of data whether
It is reported according to the renewal frequency of regulation；Specifically, timeliness is defined as:
Wherein, S is timeliness, tdFor the lag time that timeliness constraint rule defines,For average report cycle, formula are as follows:
Wherein, tinFor data inputting time, toccurFor data time of origin, NrecordFor logging data amount, p is data inputting
Number, q are the number of data set.
6. tobacco industry batch overall process quality testing as described in claim 1 and improved method, which is characterized in that
The uniqueness refers in across enterprise batch management process that data cannot repeat to upload and be recorded, and needs to guarantee to upload data
It is unique；Specifically, unique definition are as follows:
Wherein, R1For uniqueness, NallFor all Mission Numbers, NrepeatFor duplicate Mission Number.
7. tobacco industry batch overall process quality testing as described in claim 1 and improved method, which is characterized in that
The validity refers to that data are effective in threshold range, and is invalid or wrong beyond the data after threshold value；
Specifically, validity is defined as:
Wherein, V is validity, and the data that x (i) is i-th, S is the total amount of data,xWithRespectively under the threshold value of data area
Limit and the upper limit.
8. tobacco industry batch overall process quality testing as described in claim 1 and improved method, which is characterized in that
The stability refers in the production process of enterprise, reflects fluctuation situation of the process data in threshold range；Specifically, surely
Qualitative definition are as follows:
Wherein,xWithThe respectively bottom threshold and the upper limit of data area.
9. tobacco industry batch overall process quality testing as described in any one of claims 1 to 8 and improved method,
It is characterized in that, the calculating process of the overall target are as follows:
(a) by integrality, accuracy, consistency, timeliness, uniqueness, validity and 7 class data evaluation index of stability into
Row forward directionization processing；
(b) it is compared scale using significance level of the method for expert estimation to data evaluation index, constructs one and judges square
Battle array A (aij)n×n, judgment matrix A (aij)n×nIn, the importance value of each element representation data evaluation index, and each element is closed
In diagonal line at reciprocal relation；
(c) judge judgment matrix A (aij)n×nMaximum eigenvalue λmaxCorresponding feature vector, then normalizing is carried out to feature vector
Change processing, obtained vector is the subjective weight vectors w=(ω of each data evaluation index1,...,ωn)；
(d) the entropy weight vector for calculating every class data evaluation index, specifically, for individual data evaluation index, comentropy are as follows:
Wherein,M is all quantity for participating in evaluation, yijFor i-th of data evaluation index of j-th of unit；
The entropy weight weight ε of i-th of indexi:
(e) be directed to each data evaluation index, by step (c) obtain subjective weight and step (d) obtain entropy weight weight into
Row combination, obtains the comprehensive weight of each data evaluation index；
(f) the comprehensive evaluation index F of data is calculated according to the comprehensive weight of data evaluation indexj:
Wherein, yijFor i-th of data evaluation index of j-th of unit, αiFor the comprehensive weight of i-th of data evaluation index.
10. a kind of tobacco industry batch overall process quality testing and improved device characterized by comprising
Data acquisition module: for acquiring production environment data, production process data, production equipment technological parameter data, production
Process quality detecting data；
Data quality accessment module: being directed to every class data, calculates at least one data evaluation index corresponding with every class data,
Middle data evaluation index includes integrality, accuracy, consistency, timeliness, uniqueness, validity and the stability 7 of data
Class data evaluation index, and the overall target for acquiring data is calculated according to 7 class data evaluation indexs；
Quality of data diagnosis and improve module: for when aggregation of data index is unsatisfactory for healthiness condition, then according to aggregation of data
Index reversely deduces the data evaluation index of problem, and improves mechanism according to data to that class data evaluation index to go wrong
Corresponding data improve amendment, wherein data improve mechanism include: missing data amended record and passback, repeated data
Cleaning and the warning not in time of check, the rejecting of invalid data, the amendment of logic error, the processing of inconsistent data, data, number
According to the limitation of wide fluctuations.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910322172.3A CN110033201A (en) | 2019-04-22 | 2019-04-22 | A kind of tobacco industry batch overall process quality testing and improved method and apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910322172.3A CN110033201A (en) | 2019-04-22 | 2019-04-22 | A kind of tobacco industry batch overall process quality testing and improved method and apparatus |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110033201A true CN110033201A (en) | 2019-07-19 |
Family
ID=67239425
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910322172.3A Pending CN110033201A (en) | 2019-04-22 | 2019-04-22 | A kind of tobacco industry batch overall process quality testing and improved method and apparatus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110033201A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111310803A (en) * | 2020-01-20 | 2020-06-19 | 江苏神彩科技股份有限公司 | Environment data processing method and device |
| CN111414684A (en) * | 2020-03-16 | 2020-07-14 | 中国农业大学 | Method, device and system for controlling packaging quality and computer readable storage medium |
| CN111950785A (en) * | 2020-08-05 | 2020-11-17 | 上海微亿智造科技有限公司 | Event prediction and factor recognition dynamic simulation model modeling method and system |
| CN112794022A (en) * | 2021-01-26 | 2021-05-14 | 红云红河烟草(集团)有限责任公司 | Cigarette carton flow monitoring system |
| CN114205259A (en) * | 2021-12-07 | 2022-03-18 | 施耐德电气（中国）有限公司 | Method and device for diagnosing abnormal counting of gateways |
| CN115392811A (en) * | 2022-10-31 | 2022-11-25 | 成都飞机工业（集团）有限责任公司 | Method, device and equipment for evaluating quality of main data and storage medium |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101894319A (en) * | 2010-06-28 | 2010-11-24 | 中国烟草总公司湖南省公司 | Tobacco enterprise data quality management system and method |
| CN102902882A (en) * | 2012-09-14 | 2013-01-30 | 安徽南瑞继远软件有限公司 | Method for evaluating operation quality of information systems |
| CN103247008A (en) * | 2013-05-07 | 2013-08-14 | 国家电网公司 | Quality evaluation method of electricity statistical index data |
| CN104794206A (en) * | 2015-04-23 | 2015-07-22 | 国网山东省电力公司 | Transformer substation data quality evaluation system and method |
| US20170323233A1 (en) * | 2016-05-09 | 2017-11-09 | Mighty Al, Inc. | Tasking system |
| CN107463532A (en) * | 2017-06-28 | 2017-12-12 | 国网上海市电力公司 | A kind of mass analysis method of electric power statistics |
| CN107491381A (en) * | 2017-07-04 | 2017-12-19 | 广西电网有限责任公司电力科学研究院 | A kind of equipment condition monitoring quality of data evaluating system |
| CN107862456A (en) * | 2017-11-02 | 2018-03-30 | 江苏中烟工业有限责任公司 | A kind of method of batch quality of tobacco overall merit |
| CN108038626A (en) * | 2017-12-27 | 2018-05-15 | 山东中烟工业有限责任公司 | A kind of cigarette machine Life cycle monitoring system and monitoring method |
| CN108681814A (en) * | 2018-05-10 | 2018-10-19 | 北京鼎泰智源科技有限公司 | A kind of big data quality standard management control method |
| EP3435303A1 (en) * | 2017-07-25 | 2019-01-30 | Siemens AG Österreich | Method for using signifiers for data integration in rail automation |
| CN109426225A (en) * | 2017-09-01 | 2019-03-05 | 秦皇岛烟草机械有限责任公司 | A kind of method of quality control of tobacco cutting production |
-
2019
- 2019-04-22 CN CN201910322172.3A patent/CN110033201A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101894319A (en) * | 2010-06-28 | 2010-11-24 | 中国烟草总公司湖南省公司 | Tobacco enterprise data quality management system and method |
| CN102902882A (en) * | 2012-09-14 | 2013-01-30 | 安徽南瑞继远软件有限公司 | Method for evaluating operation quality of information systems |
| CN103247008A (en) * | 2013-05-07 | 2013-08-14 | 国家电网公司 | Quality evaluation method of electricity statistical index data |
| CN104794206A (en) * | 2015-04-23 | 2015-07-22 | 国网山东省电力公司 | Transformer substation data quality evaluation system and method |
| US20170323233A1 (en) * | 2016-05-09 | 2017-11-09 | Mighty Al, Inc. | Tasking system |
| CN107463532A (en) * | 2017-06-28 | 2017-12-12 | 国网上海市电力公司 | A kind of mass analysis method of electric power statistics |
| CN107491381A (en) * | 2017-07-04 | 2017-12-19 | 广西电网有限责任公司电力科学研究院 | A kind of equipment condition monitoring quality of data evaluating system |
| EP3435303A1 (en) * | 2017-07-25 | 2019-01-30 | Siemens AG Österreich | Method for using signifiers for data integration in rail automation |
| CN109426225A (en) * | 2017-09-01 | 2019-03-05 | 秦皇岛烟草机械有限责任公司 | A kind of method of quality control of tobacco cutting production |
| CN107862456A (en) * | 2017-11-02 | 2018-03-30 | 江苏中烟工业有限责任公司 | A kind of method of batch quality of tobacco overall merit |
| CN108038626A (en) * | 2017-12-27 | 2018-05-15 | 山东中烟工业有限责任公司 | A kind of cigarette machine Life cycle monitoring system and monitoring method |
| CN108681814A (en) * | 2018-05-10 | 2018-10-19 | 北京鼎泰智源科技有限公司 | A kind of big data quality standard management control method |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111310803A (en) * | 2020-01-20 | 2020-06-19 | 江苏神彩科技股份有限公司 | Environment data processing method and device |
| CN111414684A (en) * | 2020-03-16 | 2020-07-14 | 中国农业大学 | Method, device and system for controlling packaging quality and computer readable storage medium |
| CN111950785A (en) * | 2020-08-05 | 2020-11-17 | 上海微亿智造科技有限公司 | Event prediction and factor recognition dynamic simulation model modeling method and system |
| CN112794022A (en) * | 2021-01-26 | 2021-05-14 | 红云红河烟草(集团)有限责任公司 | Cigarette carton flow monitoring system |
| CN112794022B (en) * | 2021-01-26 | 2024-03-12 | 红云红河烟草(集团)有限责任公司 | Cigarette flow monitoring system |
| CN114205259A (en) * | 2021-12-07 | 2022-03-18 | 施耐德电气（中国）有限公司 | Method and device for diagnosing abnormal counting of gateways |
| CN115392811A (en) * | 2022-10-31 | 2022-11-25 | 成都飞机工业（集团）有限责任公司 | Method, device and equipment for evaluating quality of main data and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110033201A (en) | A kind of tobacco industry batch overall process quality testing and improved method and apparatus | |
| CN109784806B (en) | Supply chain control method, system and storage medium | |
| Ban et al. | Performance evaluation model of Romanian manufacturing listed companies by fuzzy AHP and TOPSIS | |
| Fang et al. | Optimization for a three-stage production system in the Internet of Things: procurement, production and product recovery, and acquisition | |
| Bartelsman et al. | Cross-country differences in productivity: The role of allocation and selection | |
| Koren et al. | Collaborative filtering on ordinal user feedback | |
| CN109408359A (en) | A kind of software test procedure quality metric method and system | |
| Pfleeger et al. | Software metrics in the process maturity framework | |
| CN105867341B (en) | A kind of the online equipment health status self checking method and system of tobacco processing equipment | |
| Wang | Applying 2-tuple multigranularity linguistic variables to determine the supply performance in dynamic environment based on product-oriented strategy | |
| CN105334185A (en) | Spectrum projection discrimination-based near infrared model maintenance method | |
| CN110728422A (en) | Building information model, method, device and settlement system for construction project | |
| CN109034483A (en) | A kind of inspection planning method based on quality function deploying | |
| Jamshi et al. | Causal linkage among business analytics, supply chain performance, firm performance and competitive advantage | |
| Esmaeili et al. | A game theory model for pricing and supplier selection in a closed-loop supply chain | |
| Putri et al. | The concept of supply chain management performance measurement with the supply chain operation reference model (Journal review) | |
| Soo | Intra‐industry trade: A Krugman–Ricardo Model and Data | |
| CN111932044A (en) | Steel product price prediction system and method based on machine learning | |
| Taghavi et al. | Investigation of the effect of innovative management and strategic marketing on the export performance of greenhouse grown crops (case study: greenhouses of Varamin county) | |
| CN102156916A (en) | Method for assessing reliability process capability of aviation product research unit | |
| CN113742248A (en) | Method and system for predicting organization process based on project measurement data | |
| CN110838006B (en) | Test system for universal device | |
| CN110288265B (en) | Multi-element key quality characteristic coupling analysis method based on modularization | |
| CN111849545B (en) | Hydrocracking product quality prediction method, device and memory | |
| CN112396457A (en) | Smart supply chain big data sharing method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190719 |
|
| RJ01 | Rejection of invention patent application after publication |