EP2352867B1

EP2352867B1 - A method for optimizing a manufacturing process in a textile plant

Info

Publication number: EP2352867B1
Application number: EP09767925.2A
Authority: EP
Inventors: Flavio Carraro; Chandran Prabakaran; Krishnan Muraliganesh; Hansruedi Wampfler
Original assignee: Uster Technologies AG
Current assignee: Uster Technologies AG
Priority date: 2008-11-14
Filing date: 2009-11-13
Publication date: 2014-04-16
Anticipated expiration: 2029-11-13
Also published as: EP2352867A1; JP2012508837A; WO2010054497A1; CN102216503A; CN102216503B; JP5382475B2

Description

BACKGROUND OF THE INVENTION

The invention relates to a method for optimizing with regard to quality, productivity and/or profitability a manufacturing process in a textile plant such as a spinning mill, weaving mill or embroidery plant, according to the preamble of the first claim.

DESCRIPTION OF THE PRIOR ART

Various raw materials are processed in spinning mills in several processing steps via intermediate products into yarns as end products. The raw materials pass through various work stations such as blowing, opening, cleaning, mixing, carding, combing, drafting, roving, fine spinning and finally spooling and winding. Machines are used for most steps which are equipped with sensors. The sensor signals are used for controlling the processing and/or for monitoring the quality of the produced intermediate and end products.
A method and a system for quality monitoring in a spinning mill are known from DE-41' 13'384 C2 , in which the data collected for a lot or batch in a single instance are linked with this lot of the processed material and are stored centrally, so that the passage of material on the basis of a lot is completely traceable and interventions can be made in the production process of an end product in the case of quality deficiencies. This procedure aims at tracing more rapidly the errors in a processing chain having several stations. The detected data are also saved, so that correlations of problems influencing the quality especially over various batches can be determined. The goal of DE-41'13'384 C2 is quality assurance by monitoring the machines used in the processing chain.
EP-0'365'901 A2 relates to a system for monitoring a plurality of textile machine workstations, e.g., winding positions of an automatic winding machine. The system contains measurement elements associated with the workstations, and means for evaluating the signals supplied by the measurement elements. Characteristic parameters are obtained during the evaluation for the individual workstations and analyzed for significant deviations from the corresponding desired values. The desired values are formed from the behavior of a statistically comparable collective. At the beginning of each monitoring operation, generalized start values are used for the individual desired values, which are converted during the course of the monitoring into more accurate, absolute values. These are updated continuously and form the core data for an automatic inference process. Consequently, the evaluation of the measurement results becomes independent of the interpretation of the operating personnel.
EP-0'410'429 A1 discloses a method and an apparatus for operating a spinning mill comprising various regions such as blow room, spinning preparation and spinning. One measures at least one quality feature of the respectively produced fiber structure at at least some textile machines in each of the regions and uses it to regulate the respective textile machine or an earlier textile machine of the same region. Correction values are formed from at least some of the quality features measured in the spinning preparation and spinning regions and are used to influence the operation of a textile machine of an earlier region.
US-2005/0159835 A1 describes a quality assurance model in which process status data and product check data are determined during a production process. The process status data concern data which are obtained continuously from the process steps during production, and the product check data are associated with the semi-finished goods or end products produced after a process. A characteristic quantity of these data is taken for each product or product group. Such characteristic data quantities and the process status data are then placed in correlation, and the quality assurance model is thus generated in such a way that an analysis is performed by data mining, such that the data placed in correlation are processed, so that a relationship is obtained between the data to the characteristic data quantities and the process status data. In other words, this specification discloses a datamining method for optimizing the process steps during the production process, which is disclosed in the sequence of different process steps in semiconductor production. Based on this technology, the same end products are assumed and the difficulties lie in managing the process and the boundary conditions there. This is a different initial situation than in the textile industry.

SUMMARY OF THE INVENTION

In addition to quality assurance, a spinning mill also has an interest to know the quality, profitability and productivity to be achieved on the basis of purchased batches and lots. This knowledge is essential for the spinning mill because the measurable properties of the yarns, which can also be designated as yarn quality, is the relevant quantity for the target agreements with the industry that further processes the yarns or with the merchants purchasing the yarns.
Such target agreements can be found in the delivery agreements and the underlying physical or chemical properties are checked upon delivery of the merchandise. In the textile industry, the tabular values of "USTER^® STATISTICS" of Uster Technologies AG, Uster, Switzerland, are frequently used for such agreements. The "USTER^® STATISTICS" concern quality reference data relating to worldwide textile production. They can be retrieved from the website www.uster.com or ordered from Uster Technologies AG, 8610 Uster, Switzerland. In order to compile them, a plurality of textile quality parameters of the participants on the market are measured over a relevant past period of time and their statistical distributions are published. They are used as a benchmark tool for determining and agreeing upon quality properties of textile structures such as yarns on the basis of their respectively measured parameters. In the event of non-adherence to such agreed target requirements, the yarns supplied by the spinning mill will not be accepted for example, will be returned or price reductions will be demanded. That is why spinning mills mostly use batches for processing an order in which the spinning mills are sure that the yarn produced therefrom will reliably meet the agreed target demands.
It is disadvantageous in this case that for reliably achieving the physical yarn parameters it is necessary to purchase raw material batches which clearly exceed the mentioned parameters, which usually entails a disadvantage in regard to pricing.
Moreover, the process parameters of the employed machines are set conservatively, so that the productivity, i.e., the production of yarn per unit of time, will decrease.
It is an object of the invention to optimize a textile manufacturing process with regard to quality, productivity and/or profitability. A method shall be provided in which the employed qualities of a lot are as close as possible to the quality of a lot which can also be designated as necessary quality of a lot in order to offer the yarn quality provided for a delivery agreement. It is a further object of the invention to also enable such an evaluation when such measured data are combined with each other only subsequently because the individual production stations are operated in an isolated manner with respect to the transducers and data handover. It is a still further object of the invention to alert a user in case of abnormal events or states.
These and other objects are solved by the method and the computer program product as defined in the independent claims. Advantageous embodiments are indicated in the dependent claims.
The invention is based on the finding that the data collected with the individual sensors in the various steps of the processing chain will allow drawing conclusions on the resulting yarn qualities and the process productivity and profitability. The measured data of different completed processing steps with their defined material flows are linked together, such that measured parameters are acquired in at least two different processing steps, and such values are stored in a database and linked with each other in an index file. Thus, gaps between different processing steps are bridged. Mutual influences and dependencies are represented in a model of the manufacturing process.
As far as possible, all data available from the whole manufacturing process shall be linked with each other in the index file, in order to completely represent the manufacturing process. The data include material data, processing data and data related to time. In practice, however, there may be cases in which a complete data collection and linking is not possible. Some data may not be available, for instance if some machines are not equipped with sensors or if some sensors are not working. Therefore, the evaluation of the data should be tolerant with respect to missing input information.
In the method for optimizing with regard to quality, productivity and/or profitability a manufacturing process in a textile plant according to the invention, raw material is processed in a manufacturing process in several processing steps into intermediate products and an end product is produced. Parameters of the raw material, the intermediate products and/or the end product are measured and stored in a database. The parameters are measured in at least two different processing steps, stored in the database and linked in an index file.
According to an alternative related to forward engineering, the data stored in the database are statistically evaluated, at least one parameter of the raw material is predetermined and at least one parameter of the end product is determined by comparison of the data stored in the database with the at least one predetermined parameter of the raw material, depending on a chosen manufacturing process.
According to an alternative related to backward engineering, the data stored in the database are statistically evaluated, at least one parameter of the end product to be produced is predetermined and at least one parameter of the raw material is determined by comparison of the data stored in the database with the at least one predetermined parameter of the end product, depending on a chosen manufacturing processes.
In a preferred embodiment, the at least two different processing steps are performed on at least two different work stations or machines. At least one parameter can be measured for the raw material used in the processing steps and/or at least one parameter can be measured for the end product produced in the manufacturing process. In addition to the measured parameters, parameters of the manufacturing process can additionally be stored in the database and linked in the index file. Such parameters of the manufacturing process comprise, e.g., a setting of a machine, a characteristic of a machine and/or a temporal workflow.
The statistical evaluation of the data stored in the database is preferably performed in an evaluation module. The statistical evaluation may comprise data mining.
According to a preferred embodiment of the method according to the invention, in a first step the parameters of the raw material, the parameters measured in the processing steps and the parameters of the end product are displayed in a display and input module, so that an index page for the index file is entered in order to manually link the data associated with the said manufacturing process in the index file. The displaying may comprise the displaying of the parameters from predetermined time window, which are associated in respect of time to a manufacturing process.
An alarm can be triggered when an abnormal event or state is detected based on the data stored in the database. The threshold parameter values for triggering the alarm can be set by a user or, by a manufacturer or automatically calculated from other data.
The computer program product according to the invention comprises a program code stored on a machine-readable carrier for executing the method according to the invention as described above when the program product runs on a computer.
In summary, two new options are provided to the textile expert with the use of the invention:

Forward engineering: On the one hand, the textile expert can determine the achievable quality and productivity of end products from the measured data of raw materials, semi-finished goods and/or processes, in that he or she makes an ongoing comparison between such current raw fibers and the semi-finished goods with data from earlier productions.
Backward engineering: On the other hand, the textile expert can delimit the permissible characteristics and properties of raw materials, semi-finished goods and/or processes which can be tolerated for a known or new finished product, which occurs on the basis of comparison with the same or similar finished products from earlier production processes.

It is further possible that the evaluation module provides proposals for setting machines in the production chain or changes these settings in the machine directly, if the evaluation module is connected with the machines.
In the case of businesses that operate several spinning mills, these data can be connected throughout the group in order to optimize the product portfolio. Orders can then be awarded depending on how the productivity and the quality differ between the various spinning mills for the same or similar products.
Although the invention is described subsequently by reference to a spinning mill, this method can also be employed generally in other textile plants such as weaving mills or embroidery plants.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention is now explained in closer detail by references to drawings which illustrate an embodiment of the invention.

Figure 1: shows a graphical representation of the values of a quality parameter Q in a comparison of conventional values according to the state of the art with the values to be achieved with the invention.
Figure 2: shows a material flow diagram, with a selection of manufacturing processes being shown.
Figure 3: schematically shows a database and an evaluation module according to the invention.
Figure 4: schematically shows an allocation of an end product to a raw material by a process.
Figure 5: shows a graphical illustration of two values of a parameter and the allocation of products under these parameters in accordance with the invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

Figure 1 shows a diagram with a possible target agreement in the form of a predetermined quality curve 103 for a textile structure such as a yarn. A specific imperfection parameter such as the size (e.g., diameter, or a product of diameter and length) of the imperfection is entered along the horizontal axis 101 and a frequency of the respective imperfection parameter such as the number of imperfections is entered along the vertical axis 102. Individual yarns are represented by measuring points. A first yarn 104, which is illustrated by its measuring point arranged above the curve 103, would be regarded as non-satisfactory with reference to this imperfection parameter. When using a device that reduces the imperfections such as a yarn clearer, the quality of the yarn can be improved to quality 114. The parameterization of the device is relevant, as is shown by arrow 124, which therefore cuts out a higher number of imperfections. This is at the expense of the productivity (speed of the yarn passage) however.
According to the state of the art, lots or batches are used for production in which respective measured values 105, 106, 107 for the produced yarns are obtained, either by selection of the respective batches or the setting of processing stations that influence the quality or both. Yarns are thus obtained which more than exceed the target agreement with the purchaser, which can be seen from the fact that the measuring points 105, 106 and 107 maintain a large distance from the target curve 103. This is at the expense of processing speed and/or increases the costs for the selection of the raw materials.
The invention now allows the selection of the parameter settings in the individual processing stations on the one hand and also the selection of the suitable raw product batches, e.g., the cotton bales, on the other hand, which occurs in such a way that the measured values 115, 116, 117 are much closer to the target curve 103, leading to cost benefits for the spinning mill and ensuring at the same time the fulfillment of the target agreements.
Notice must be taken in this connection that the agreements of spinning mills with the purchasers are linked in respect of quality to a plurality of quality parameters. This means for example that a lower number of imperfections or slubs do not necessarily mean a higher quality for the purchaser. A higher number of imperfections or slubs can be more inconspicuous at higher roughness of the yarn than a low number.
In this respect, a quality vector which characterizes a yarn and is the result of the plurality of measured values determining the quality properties is relevant in quality evaluation.
Notice must further be taken that the quality (of products and processes) is not the only criterion that needs to be considered in a textile plant. Further important criteria are profitability (of products) and productivity (of processes).
It is an advantage of the invention that a plurality of parameters can be considered. Figure 2 shows a material flow diagram, with a selection of manufacturing processes being shown. The input side shows raw materials 201 such as textile fibers for example which can be characterized by different provenances, and in the case of the same provenance by individual batches with different properties. Fiber mixtures can also be used. That is why a combination of two different raw materials 201 is shown by the two arrows 211 which then pass through a predetermined manufacturing step 202. The different manufacturing processes depend on the textile businesses and can comprise for example the processes of opening, cleaning, mixing, carding, combing, drafting, spinning and finally winding on a body such as a bobbin for further use, especially for delivery to a purchaser. These process steps are characterized by the successively running intermediate processes 202 and 203, with the pictograms arranged in Figure 2 indicating which possible process routes can be taken. There are other parameters and measured values depending on the choice of the next step, indicated by arrow 212, which converges from the second to the third intermediate process. Finally, a yarn is produced which can be designated as an article which is characterized by the raw materials 201 and/or preceding work steps 202, 203. Such an end product, crystallized in a box 204, should also fulfill delivery requests made beforehand. The entire manufacturing process is characterized by the arrow 213.
In the state of the art according to DE-41'13'384 C2, individual measuring data in this material flow are used in order to identify erroneous functions and to especially identify machine processes having a negative influence on the quality. There is no influence on the parameters to achieve various qualities. In US-2005/0159835 A1 , efforts are made to increase the quality, insofar as the processes there can be applied to a spinning machine, such that optimization is carried out in adjustable parameters of the machines used in the processing steps. This procedure which is useful in physicochemical processes fails already as a result of the large number of products 204 to be obtained, because in the state of the art the process quality is aimed at obtaining a single product. Moreover, there is no consideration of the starting materials 201 in the state of the art.
Figure 3 schematically shows a database 300 which comprises data entries 301, 302, 303, 304, ..., 311, 312, 313, ..., 391, 392,393, ... and an evaluation module 3100 which carries out several functions which are described below.
The data 301 to 304 shown as pages are for instance data of raw materials 201 (see Figure 2), which are entered prior to or upon arrival at the spinning mill. This includes for instance data of cotton bales, their provenance, supplier and further data characterizing the same like degree of ripeness, etc. An arrow T indicates that these data are entered in a continuous way over time and a large number of comparative data of the raw materials 201 are present after a certain time.
Individual measurment data and/or setting data 311, 312, 313 of running intermediate processes 202, 203 (see Figure 2) are entered and stored at other places of the database 300. One example for such an intermediate process is the carding of fibers into a sliver on a carding machine.
The scope and degree of detail of the data depend strongly on the spinning mill's infrastructure. The pages of the measured values 391, 392, 393 of the various end products 204 (see Figure 2) are shown in the illustrated example as a completion.
Each of these data pages 301, 302, 303, 304..., 311, 312, 313, ... 391, 392, 393... comprises one or several measured values and thus a data vector. ,
The statistical evaluation occurs by the evaluation module 3100. In an automatic connection of the measured data of the individual steps of the flow of material, as is possible for example by an apparatus according to DE-41'13'384 C2, the module 3100 knows which starting materials 201, which are characterized here by data 301 and 303, have passed through which manufacturing process (here: 312) in order to obtain which end product 204, which is characterized here by the data 393. These data pages are each shown in Figure 3 in a hatched way. They are associated with a specific manufacturing process or flow of materials, as is indicated in Figure 2 with the arrow 213.
The indexing of the concatenated elements of a manufacturing process lead to an index file with the entries 3001, 3002, 3003, ..., with the entry 3002 comprising the reference to the elements 301, 303, 312 and 393. This entry 3002 is thus the visualization of all parameters of the manufacturing process 213 of Figure 2.
Producing this indexed connection may possibly need to be made manually because many spinning mills run individual machines that are not networked, so that the data need to be loaded into the database 300. Furthermore, the individual processing steps of an individually regarded manufacturing process characterized by process arrow 213 in Figure 2 run in a temporally successive manner, but not necessarily sequentially on the individual machines in a spinning mill, so that a direct temporal allocation is not necessarily correct. As a result, several days may lie between the start of processing and the entry of a first data page 301 on the one hand and the finishing of the product 204 on the other hand, which is then followed by preparing a data page 393.
If the machines and their sensors are networked with the database 300 in the spinning mill, the respective data pages will advantageously automatically be linked to the index pages 3001, 3002, 3003.
If the machines are not networked in the spinning mill however, the evaluation module 3100 also has the function of showing the textile experts in a spinning mill in a display and input module 3200 a suitable time window of the output data and measured data of the individual flows of material, so that the correct index page (which in this case is 3002) is produced by manual choice. The time factor of the display is important, so that temporally older data pages of prior to 1 to 3 days before the current point in time are displayed in the processed bales (the starting materials 201) and later points in time of between 2 days to 8 days before the current point in time are displayed for the processing steps of the flow of material and 16 hours to the current point in time are displayed for the end products for example. The index entry is produced by making manual selections from the displayed data pages that are not yet allocated and the allocated data pages are removed from the display. It is common practice to provide process codes instead of information on the parameters.
It can be preceded for example as follows for preparing the data pages 301, 303 in cotton bales as starting material. 17 measured values are made and four values of samples are taken for each group of bales. Mean values are then formed and said 17 mean values are recorded for the data page 301. Depending on the configuration of the spinning mill, the number of data values can be lower or higher. The mean value formation also depends on the quantity and the expected spreading of the output material.
Figure 4 shows two spaces: A first space 401 which is characterized by parameters x₁, x₂ of raw materials, as are designated in Figure 2 with reference numerals 201, and a second space 402 which is characterized by parameters y₁, y₂ of end products, as are designated in Figure 2 with reference numeral 204. Points 411, 412 are entered in the two spaces 401, 402 schematically, which points correspond to certain measured raw materials 201 and end products 204, respectively. The parameters y₁, y₂ of end products 204 can be combined if required into a single quality parameter. An arrow 403 in Figure 4 which allocates a raw material 201 to an end product 204 represents a specific manufacturing process P. This fact can be expressed as follows in a symbolic way: $P : x_{1}, x_{2} \to y_{1} (x_{1}, x_{2}), y_{2} (x_{1}, x_{2}) .$
In order to assess the effects of the changes Δx₁, Δx₂ of the parameters x₁, x₂ of the raw materials 201 on the parameters y₁, y₂ of the end products 204, it is advantageous to know the variables for as many processes P which correspond at least approximately to the partial derivatives $\frac{\partial y_{1}}{\partial x_{1}}, \frac{\partial y_{1}}{\partial x_{2}}, \frac{\partial y_{2}}{\partial x_{1}}, \frac{\partial y_{2}}{\partial x_{2}} .$
They can be approximated numerically if parameters y₁, y₂ of the end products 204 are known for the respective process P, which correspond to such raw materials 201 whose parameters x₁, x₂ are as close as possible to the respective point 411. In the case of known partial differentiations, the changes Δy₁, Δy₂ of the parameters of the end products 204 can be stated as follows: $Δ y_{1} = \frac{\partial y_{1}}{\partial x_{1}} Δ x_{1} + \frac{\partial y_{1}}{\partial x_{2}} Δ x_{2}; Δ y_{2} = \frac{\partial y_{2}}{\partial x_{1}} Δ x_{1} + \frac{\partial y_{2}}{\partial x_{2}} Δ x_{2} .$
It is also advantageous to define certain "standard processes" with which the respective textile plant will work relatively frequently and of which there are consequently a relative large number of measured data. Small deviations from such a standard process such as the use of another machine for a certain intermediate process 202, 203 can be handled in a computational respect very easily, e.g. in the form of a corrective factor which can each be determined for the respective machine and a specific parameter y₁, y₂ of the end products 204.
It is understood that the two spaces 401, 402, which are shown in Figure 4 in two dimensions for reasons of simplicity, can have any desired number of dimensions. In order to reduce the dimensions and/or in order to ensure that the parameters will become linearly independent of each other, methods known from multi-variant statistics can be employed, e.g., principal component analysis (PCA).
The allocation 403 does not have to be unique, as is illustrated with the exemplary process 213 of Figure 2. If this is the case and/or if the data in the end product space 402 lie far apart, known interpolation or extrapolation methods can be applied, see for example J. Gleue, "Triangulierung und Interpolation von im R² unregelmäßig verteilten Daten" (Triangulation and Interpolation of Data Distributed Irregulary in R²), Hahn-Meitner-Institut für Kernforschung Berlin GmbH, Report No. HMI-B 357, July 1981. Further known possibilities for easier handling of the data in a computational respect are the following: data reduction, data combination, elimination of outliers, resampling and/or smoothing.
Figure 5 shows a graphical display of two values 501 and 502 of a parameter of the property of a yarn as an end product and the allocation of products 511, 512 according to the invention and a product 513 according to the state of the art. They concern threshold values which are to be undercut by the respective values from the data pages 391, 392, 393.
A vertical axis 500, which is marked with %, shows a random parameter for example from the large number of the property values for yarns as predetermined by the benchmark tool USTER^® STATISTICS. A statement of 50 %, as in the value 501 for example, means that 50 % of all materials tested in preparation of the benchmark tool had this value or better and the other 50 % had a worse value. With the value 502, the 25 % value can be stated for the same parameter for example, wherein in this case only 25 % of all tested materials have this value or better. With suitable raw materials which meet the 50 % value, an area 513 can be covered with a certain yarn thickness according to the state of the art. Sections of the area closer to the marking 501 closely meet this parameter, and sections further away from the marking 501 more than exceed this parameter. By detecting and evaluating different flows of materials by the method of data mining for specific raw materials and specific processing steps, this variance of the parameter inspected here can be narrowed down. Variance can be reduced for example by a factor 2, so that by choosing a certain flow of material which also includes the choice of certain raw materials and optionally settings of machines of the flow of material, a yarn can be produced which in respect of this parameter has a higher quality and remains below the better benchmark 502. With other parts of the flow of material and raw material the product can be achieved with the specifications of value 501, but with a lower variance, so that higher value creation is possible by combination of the values because the agreed default parameters are not undercut too excessively.
For the spinning mill, the order is relevant which defines a predetermined number of parameters of the product to be delivered. These parameters are reflected directly or approximately by the data pages 391, 392, ... (cf. Figure 3). By linking the data of the processing steps 202, 203 under inclusion of the raw materials 201, it is thus possible with the statistical methods of data mining in the evaluation module 3100 to provide an output of a possible flow of material (not shown in the drawings, but corresponding to Figure 3), which output undercuts the defaults of the product to be supplied only by measurement precision and the variance of all predetermined parameters. This technical decision of the choice of a computable set of data on a flow of material can also comprise a compilation of individual data which are not yet saved as a data record in this form and form a new data record 3008 after the completion of the production lot. In this case, it is not the order parameters but the end product defined by the new material flow that is characterized on the basis of its parameterized properties.
In this process, measured data that cannot be influenced directly in the evaluation, such as quality data of the starting materials, are placed in correlation with the setting data and the measured data of the existing machines, in order to limit the scattering of the large number of theoretical end products obtained by the processing. Thus the quality parameters of the end products come as close as possible to and are slightly better than the agreed delivery data, which allows an especially temporal improvement of the production flows in the setting data of the existing machines.
In addition to the backward engineering with the evaluation as explained above, it is also possible by using methods of data mining to provide parameterizations of machines of the flow of material at given starting materials, which are characterized by data pages 301, etc., in order to also achieve predetermined quality parameters of end products, which therefore means performing forward engineering.

LIST OF REFERENCE NUMERALS

101: Axis of defects
102: Axis of frequency
103: Quality curve
104: Yarn, lack of quality
105, 106, 107: Measured values (state of the art)
114: Yarn, improved quality
115, 116, 117: Measured values (invention)
124: Reduction in defects

201: Raw material
202, 203: Processing steps
204: End product
211: Combination of two different raw materials
212: Transition to next step in processing
213: Manufacturing process

300: Database
301, 302, 303, 304: Data pages for starting material
311, 312, 313: Data pages for measured values of manufacturing process
391, 392, 393: Data pages for measured values of end product
3001, 3002, 3003: Index files of data pages
3008: New data record
3100: Evaluation module
3200: Display and input module

401: Space of raw materials
402: Space of end products
403: Allocation
411: Point of a raw material
412: Point of an end product

500: Parameter value from USTER^® STATISTICS for example
501, 502: Threshold values of the parameter
511, 512: Properties of a yarn with respect to parameter (invention)
513: Properties of a yarn with respect to parameter (state of the art)

Claims

A method for optimizing with regard to quality, productivity and/or profitability a manufacturing process (213) in a textile plant, wherein
raw material (201) is processed in the manufacturing process (213) in several processing steps (202, 203) into intermediate products and an end product (204) is produced, and
parameters (301, 303, 312, 393) of the raw material (201), the intermediate products and/or the end product (204) are measured in at least two different processing steps (202, 203),
characterized in that
the measured parameters (301, 303, 312, 393) are stored in a database (300) and linked in an index file (3002),
the data (301, 303, 312, 393, 3002) stored in the database (300) are statistically evaluated, and
at least one parameter (304) of the raw material (201) is predetermined and at least one parameter of the end product (204) is determined by comparison of the data (301, 303, 312, 393, 3002) stored in the database (300) with the at least one predetermined parameter (304) of the raw material (201), depending on a chosen manufacturing process (213), or
at least one parameter of the end product (204) to be produced is predetermined and at least one parameter of the raw material (302) is determined by comparison of the data (301, 303, 312, 393, 3002) stored in the database (300) with the at least one predetermined parameter of the end product (204), depending on a chosen manufacturing processes (213).
The method according to claim 1, wherein the at least two different processing steps (202, 203) are performed on at least two different work stations or machines.
The method according to one of the preceding claims, wherein at least one parameter (301, 303) is measured for the raw material (201) used in the processing steps (202, 203) and/or at least one parameter (393) is measured for the end product (204) produced in the manufacturing process (213).
The method according to one of the preceding claims, wherein parameters of the manufacturing process (213) are additionally stored in the database (300) and linked in the index file (3002).
The method according to claim 4, wherein the parameters of the manufacturing process (213) comprise a setting of a machine, a characteristic of a machine and/or a temporal workflow.
The method according to one of the preceding claims, wherein the statistical evaluation of the data (301, 303, 312, 393, 3002) stored in the database (300) comprises data mining.
The method according to one of the preceding claims, wherein in a first step the parameters (301, 303) of the raw material (201), the parameters (312) measured in the processing steps (202, 203) and the parameters (393) of the end product (204) are displayed in a display and input module (3200), so that an index page for the index file (3002) is entered in order to manually link the data associated with the said manufacturing process (213) in the index file (3002).
The method according to claim 7, wherein the displaying comprises the displaying of the parameters from predetermined time windows, which are associated in respect of time to a manufacturing process (213).
The method according to one of the preceding claims, wherein an alarm is triggered when an abnormal event or state is detected based on the data (301, 303, 312, 393, 3002) stored in the database (300).
A computer program product comprising a program code stored on a machine- readable carrier for executing the method according to one of the preceding claims when the program product runs on a computer.