How to Analyze Quality Defects

Abstract

In order to provide a method for analyzing quality defects of workpieces, preferably vehicle bodies and / or vehicle attachment parts, in particular after and / or as they pass through a manufacturing process in a method technology facility, preferably after and / or as they pass through a painting process in a painting facility, which method makes it possible to avoid quality defects and / or to determine, avoid and / or eliminate causes of quality defects within the manufacturing process, it is proposed according to the invention that the method comprises: - forming a workpiece-specific data set uniquely associated with the workpiece at the start of the manufacturing process, in particular at the start of the painting process, and / or forming a workpiece support-specific data set uniquely associated with the workpiece support at the start of the manufacturing process, in particular at the start of the painting process; - supplementing the workpiece-specific data set with, in particular, quality-related process data while the workpiece passes through the manufacturing process, in particular the painting process, and / or supplementing the workpiece support-specific data set with, in particular, quality-related process data while the workpiece passes through the manufacturing process, in particular the painting process; - storing the workpiece-specific data set in a database and / or storing the workpiece support-specific data set in a database.

Description

[Technical field]

[0001] The invention relates to a method for analyzing quality defects of workpieces, preferably vehicle bodies and / or vehicle mounting parts, in particular after and / or during a manufacturing process in a method technology facility, preferably after and / or during a painting process in a painting facility. [Background technology]

[0002] When buying a car, potential buyers expect high quality with regard to the paintwork of the car. In this case, the respective buyer expects, for example, that the vehicle body and / or the vehicle attachment parts do not retain paintwork defects, i.e., for example, dirt inclusions, holes, runners, scratches and / or air bubbles (boiling). The buyer further expects an invariable color progression, shade, gloss level and / or an invariable shine across the vehicle body. Also desirable is an at least approximately uniform structure of the various coating and / or paint layers and / or an at least approximately uniform layer thickness of the coating and / or paint layers. Furthermore, the paintwork of the vehicle body and / or the vehicle attachment parts must also withstand external influences, for example, UV radiation, heat, cold, rain, salt, stone chips, bird droppings, dust and / or scratches from car washes.

[0003] To be able to meet these high expectations, extensive quality controls are carried out on the vehicle body and, if defects are identified, rework is possibly carried out. In the painting process, new coatings and / or paint layers are applied in each process step, so that defects in the lower layers cannot be identified within the framework of quality controls, or at most can be identified with difficulty. Quality controls are therefore required after the application of the individual coatings and / or paint layers in order to be able to guarantee the highest possible quality of the coating. However, this is not possible due to the extremely fast production cycles in the painting plant. Summary of the Invention [Problem to be solved by the invention]

[0004] The object of the present invention is to provide a method for analyzing quality defects of workpieces, preferably vehicle bodies and / or vehicle mounting parts, in particular after and / or during a production process in a method technology facility, preferably after and / or during a painting process in a painting facility, in which quality defects can be avoided and / or causes of quality defects in the production process can be determined, avoided and / or eliminated. [Means for solving the problem]

[0005] This problem is solved by a method for analyzing quality defects of workpieces, preferably vehicle bodies and / or vehicle mounting parts, in particular after and / or during a manufacturing process in a method technology installation, preferably after and / or during a painting process in a painting installation, with the features of claim 1.

[0006] The method preferably comprises: - forming a workpiece-specific data set uniquely assigned to the workpiece at the start of the manufacturing process, in particular at the start of a painting process, and / or forming a workpiece support-specific data set uniquely assigned to the workpiece support at the start of the manufacturing process, in particular at the start of a painting process; - supplementing a workpiece-specific data set with, in particular quality-relevant process data while the workpiece passes through a manufacturing process, in particular a painting process, and / or supplementing a workpiece support-specific data set with, in particular quality-relevant process data while the workpiece support passes through a manufacturing process, in particular a painting process; - storing workpiece-specific data sets in a database and / or storing workpiece support-specific data sets in a database.

[0007] The term "particularly" is used within the framework of this specification and the accompanying claims only to describe possible optional and / or alternative features.

[0008] Within the scope of this description and the accompanying claims, workpiece-specific data sets are in particular data sets which are uniquely assigned to a workpiece.

[0009] The workpiece-specific data sets form, inter alia, a "digital workpiece" and / or a "digital image" of the respective workpiece.

[0010] A workpiece support-specific data set is, within the scope of this description and the appended claims, a data set which is uniquely associated in particular with the workpiece support and in particular with all workpieces arranged on the workpiece support, for example with the workpiece support and the vehicle attachments arranged thereon.

[0011] In this case, it is particularly conceivable that a number of workpieces, in particular vehicle attachment parts, are arranged on the workpiece support.

[0012] Workpiece supports of this kind are in particular so-called skids.

[0013] For example, it is conceivable that at the start of a manufacturing process, in particular at the start of a painting process, a workpiece-specific data set and / or a workpiece support-specific data set are automatically formed, for example by using an RFID chip arranged on the workpiece and reading the application data from the chip.

[0014] The workpiece-specific data set and / or the workpiece support-specific data set are in particular generated automatically.

[0015] In particular, it is conceivable that the workpiece-specific and / or workpiece support-specific data sets, respectively, are formed from one or more data sets from a preceding process step.

[0016] For example, it is conceivable that when forming a workpiece-specific data set, body-in-white quality data from a body-in-white of a vehicle formation is added to the workpiece-specific data.

[0017] Furthermore, the workpiece-specific data set and / or the workpiece support-specific data set are It may be advantageous to subsequently supplement this with coating quality data, for example first determined in a laboratory.

[0018] The workpiece-specific data set and / or the workpiece support-specific data set preferably comprises a workpiece-specific application data set and / or a workpiece support-specific application data set, e.g. a unique workpiece identification number, a model type of the respective workpiece and / or a color code of the color to be applied on the workpiece in the painting process. The workpiece-specific application data set and / or the workpiece support-specific application data set, for example, further comprises information on the production layers during which the workpiece and / or the workpiece support passes through the production process.

[0019] The workpiece-specific data set and / or the workpiece support-specific data set preferably further comprise a manufacturing data set, which comprises process data.

[0020] In particular the quality-related process data preferably comprise physical and / or chemical influence parameters for each coating and / or paint layer that is applied onto the workpiece in the manufacturing process.

[0021] Furthermore, it may be advantageous if the especially quality-related process data comprises body-in-white quality data, for example body-in-white quality data which relates to deformations of the workpiece and / or which comprises information regarding the surface roughness of the workpiece.

[0022] The manufacturing process, in particular the painting process, preferably comprises a number of successive manufacturing steps.

[0023] In the method embodiment, the workpiece-specific data set and / or the workpiece support-specific data set are supplemented with quality data, which preferably comprises information regarding quality defects, in particular painting defects, of the respective workpiece at the end of the manufacturing process, in particular at the end of the painting process.

[0024] Preferably, the quality data is determined within the framework of quality control in a control station, at the end of the production process by a quality inspector, for example by means of visual control, by means of an automatic quality measuring station and / or by means of an automatic error detection system.

[0025] The quality data, which supplement the workpiece-specific data set and / or the workpiece support-specific data set, comprises information in particular regarding the location, type, size and / or number of quality defects.

[0026] The location of the quality defect can be determined, for example, from the coordinates of an automatic quality measuring station and / or an automatic error detection system.

[0027] The position of the quality defect can, for example, furthermore be determined by a quality inspector and can in particular be entered manually via grid coordinates.

[0028] Quality defects, in particular coating defects, include, for example, coating errors, in particular dirt inclusions, holes, runners and / or air bubbles (boiling), different layer thicknesses, undesirable color shifts and / or deviations from predetermined process result parameters.

[0029] Processing result parameters include, for example: the thickness of the layer; the flatness of the layer; the uniformity of the layer thickness; the color and / or brightness of the layer; the hardness of the layer; the chemical composition of the coating, in particular the degree of crosslinking and / or the solvent content, and / or the degree of contamination of the coating.

[0030] The quality data, which supplement the workpiece-specific data set and / or the workpiece support-specific data set, preferably further comprises information on the target status of the workpiece, i.e. whether the workpiece has to be post-processed, what post-processing has to be performed, whether post-processing is not necessary and / or whether post-processing is not possible. As post-processing, for example, one or more of the following post-processing steps are conceivable: grinding, spot repair and / or re-running of the manufacturing process, in particular the painting process.

[0031] The quality data, which supplement the workpiece-specific data set and / or the workpiece support-specific data set, preferably comprises information as to whether post-machining has been carried out.

[0032] It is further conceivable that the quality data, which supplement the workpiece-specific data set and / or the workpiece support-specific data set, contain information about whether post-machining is not beneficial. Workpieces for which post-machining is not beneficial are particularly defective.

[0033] In one embodiment of the method, quality data of a plurality of workpiece-specific data sets and / or a plurality of workpiece support-specific data sets are automatically classified into different quality defect clusters using cluster methods in order to recognize systematic quality defects.

[0034] In the cluster method, quality data of a plurality of workpiece-specific data sets and / or a plurality of workpiece support-specific data sets are preferably classified using one or more classification rules.

[0035] Preferably the quality data of the workpiece-specific data set and / or the workpiece support-specific data set are classified into systematic and non-systematic quality defect clusters.

[0036] In particular, for example, quality defects having comparable locations, types, sizes and / or numbers are classified into one quality defect cluster.

[0037] It is further conceivable, for example, that quality defects are classified into quality defect clusters according to the time of occurrence.

[0038] To classify the quality data, preferably expert defined and / or learned classification rules are used.

[0039] The classification rules defined by the experts include, for example, a defined number of workpieces having the same and / or similar quality defects within a moving window of a defined number of workpieces and / or workpiece supports passing through the manufacturing process.

[0040] For example, if identical and / or similar quality defects are recognized in three or more of five consecutive workpieces in a manufacturing process, the quality data may be classified into a systematic quality defect cluster.

[0041] Within the scope of this description and the accompanying claims, identical quality defects in this case are in particular those in which the type, size, weight and / or location of the quality defects are equal.

[0042] Similar quality defects in this case, within the framework of this description and the accompanying claims, in particular mean that the quality defects are of the same type, although the size and / or location of the quality defects may differ.

[0043] The learned classification rules are learned from quality data of the workpiece-specific data set and / or the workpiece support-specific data set, preferably using machine learning methods.

[0044] For example, it is contemplated that the classification rules may be learned using supervised and / or unsupervised machine learning methods.

[0045] In the supervised machine learning method, the quality data of the workpiece-specific data set and / or the workpiece support-specific data set are divided into systematic and non-systematic quality defect clusters. The division is preferably performed via defined rules and / or via user feedback. By means of the supervised machine learning method, preferably quality trends or future quality defect clusters can also be recognized by suitable "labeling". In the unsupervised machine learning method, the quality data, i.e. the normal state, of the workpiece-specific data set of a workpiece without quality defects and / or of the workpiece support-specific data set of a workpiece support with a workpiece without quality defects is preferably learned. If deviations from the normal state are introduced, these deviations are recognized.

[0046] The defined and / or learned classification rules are preferably transferable to other process technology plants, in particular to other painting plants.

[0047] The classification of the quality data using the cluster method is preferably performed continuously. Alternatively, or in addition, it is envisaged that the classification of the quality data by the cluster method is performed by batch processing.

[0048] The systematic quality defect clusters are preferably further determined by means of signal processing methods.

[0049] For example, to determine systematic quality defect clusters, features from a frequency density function of the quality data may be processed using one or more signal processing methods, for example using a Fourier transform and / or using density estimation.

[0050] It may be advantageous if the systematic quality defects identified by the cluster method are presented to the user by means of a visualization, for example as real-time reports in an alarm system and / or as analytical reports in an analytical system.

[0051] For example, it is contemplated that a workpiece may be rejected from a manufacturing process based on the quality defect clusters identified using the cluster method. Furthermore, it may be advantageous for a process control of the manufacturing process to be adapted based on the quality defect clusters identified using the cluster method.

[0052] In the form of a method, quality defect causes for systematic quality defects recognized using cluster methods and classified into various quality defect clusters are automatically determined using analysis methods, preferably by analyzing, in particular, quality-related process data of a workpiece-specific data set of a workpiece having systematic quality defects of the respective quality defect cluster and / or by analyzing, in particular, quality-related process data of a workpiece support-specific data set of a workpiece support having a workpiece with systematic quality defects of the respective quality defect cluster.

[0053] Preferably, therefore, within the framework of quality control, unrecognized systematic quality defects can also be recognized by inductive deduction from the production process.

[0054] In the analysis method, quality defect causes are determined using one or more analysis rules.

[0055] To determine the quality defect causes, preferably expert defined and / or learned analysis rules are used.

[0056] The analysis rules to be learned are preferably learned using machine learning methods.

[0057] For example, it is contemplated that the analysis rules may be learned using supervised and / or unsupervised machine learning methods.

[0058] In supervised machine learning methods, the analysis rules are learned via user feedback regarding quality defect causes. In unsupervised machine learning methods, normal states of the process steps of the manufacturing process are preferably learned. Deviations from the learned normal states are recognized when they occur.

[0059] The defined and / or learned analysis rules are preferably transferable to other process technology installations, in particular other painting installations.

[0060] In one embodiment of the method, the defined quality defect causes comprise anomalies and / or deviations within the manufacturing process, in particular within the painting process.

[0061] In one embodiment of the method, an analytical method is used to determine relationships between the determined quality defect causes and the systematic quality defects classified into different quality defect clusters.

[0062] The relationships between the defined quality defect causes and the systematic quality defects classified into different quality defect clusters are preferably determined by defined and / or learned analysis rules.

[0063] The relationships between the defined quality defect causes and the systematic quality defects classified into various quality defect clusters are preferably transferable to other process technology plants, in particular painting plants.

[0064] In the method aspect, future quality defects of the workpiece are automatically inductively estimated using the determined relationship using an analytical method from one or more anomalies and / or deviations determined within an ongoing manufacturing process, preferably so that impending systematic quality defects are automatically determined using the determined relationship as the workpiece passes through the manufacturing process.

[0065] In that case, the workpieces are preferably assigned to quality defect clusters already as they pass through the manufacturing process using analysis methods.

[0066] In the method embodiment, a workpiece-specific data set and / or a workpiece support-specific data set are supplemented within each individual process step with in particular quality-relevant process data.

[0067] In the embodiment of the method, one or more of the following process parameters are used as in particular quality-relevant process data supplementing the workpiece-specific data set and / or the workpiece support-specific data set: - Exceeding target times within process steps of the manufacturing process; -Events that occur while moving through process steps in the manufacturing process; -Body-in-white quality data for each workpiece -Anomalies and / or deviations in the manufacturing process; -Weather data; -People data about people involved in a process step as they move through the process step in the manufacturing process.

[0068] To supplement the respective workpiece-specific data set and / or the respective workpiece support-specific data set, one or more of the following process parameters may be advantageously used: - Nozzle parameters in the dryer; -Paint amount; -Paint color; -Paint charge; -Robot lane trajectory -Cabin temperature inside the painting cabin; -Cabin humidity in the painting cabin; - the bath temperature and filling state in the cathodic dip coating station and in the pre-treatment station; -Transport movement of the workpiece through the cathodic dip coating station; -Concentration of chemicals in the cathode dip coating station; -Voltage in the cathode dip coating station; - Temperature and pass-through time in the dryer.

[0069] The process parameters used to supplement the respective workpiece-specific data set and / or the respective workpiece support-specific data set are preferably pre-processed.

[0070] In particular, it is conceivable that process parameters with a high variability in time, such as mechanical, hydraulic or electrical process parameters, are preprocessed, but the preprocessing of process parameters with a low variability in time, i.e. inertial process parameters, such as thermodynamic process parameters, can be omitted.

[0071] Exceeding a target time means, within the framework of this specification and the appended claims, in particular exceeding a predetermined target time within a process step when performing the respective process step for the respective workpiece.

[0072] The events comprise, for example, maintenance events, alarms from the control system of the process technical installation, information about shift changes, information about production stoppages, information about entries into the processing stations of the process technical installation, information about cleaning interruptions.

[0073] With the maintenance events to be added to the workpiece-specific data set and / or to the workpiece support-specific data set it is possible, for example, to identify systematic quality defects which occur after maintenance.

[0074] The body-in-white quality data comprises in particular information regarding the sheet quality of the workpiece or parts of the workpiece, such as the bonnet, the doors, the side parts and / or the roof.

[0075] Preferably, furthermore, one or more of the following process parameters are used to supplement the respective workpiece-specific data set and / or the respective workpiece support-specific data set: -Transfer technical information; - residence time of the workpiece in the manufacturing process; -Method "Overnight storage" of workpieces inside the technical installation.

[0076] For example, the residence time of the workpiece in the dryer is used to supplement the respective workpiece-specific data set, i.e. quality defects can be inferred, in particular from exceeding a target time for the presence of the workpiece in the dryer.

[0077] Additionally, it may be advantageous if the residence time of the workpiece support within the dryer is used to supplement each workpiece support specific data set.

[0078] A deviation in a manufacturing process, within the framework of this description and the accompanying claims, is preferably a deviation of a state from a target state or a deviation of a measured process value from a predetermined process value.

[0079] The anomalies are preferably determined automatically from deviations between actual process values ​​and target windows within a process step of the manufacturing process. Alternatively or additionally, it is conceivable that the anomalies are determined automatically from deviations between actual process values ​​of a process step of the manufacturing process and normal states learned for the respective process step using machine learning methods.

[0080] Anomalies in the manufacturing process are preferably further determined using one or more subordinate microsystem analysis systems and are in particular divided at the micro level, i.e. at the workpiece level, and / or at the macro level, i.e. at the level of the process technical equipment.

[0081] In the method aspect, the workpiece-specific data set and / or the workpiece support-specific data set are supplemented with process error data while the workpiece and / or the workpiece support passes through a process step of the manufacturing process, the process error data comprising information about anomalies and / or deviations within the process step of the manufacturing process.

[0082] The process error data is specifically an "error marker."

[0083] Preferably, not only the workpiece-specific data set of the workpiece that is located in the given processing station during the production and / or transfer stop is supplemented with the process error data, but rather, it is conceivable that the workpiece-specific data sets of other workpieces that are also located in the given processing station 104 for a longer period due to the production and / or transfer stop are also supplemented with the process error data.

[0084] It is further preferred that not only the workpiece support-specific data sets of a workpiece support which is located in a given processing station during a production and / or transport stop are supplemented with process error data, but in particular it is also conceivable that the workpiece support-specific data sets of other workpiece supports which are likewise located in a given processing station for a longer period due to a production and / or transport stop are also supplemented with process error data.

[0085] Using the process error data, anomalies and / or deviations within a process step of a manufacturing process can be preferably identified as errors.

[0086] Preferably, by using process error data to identify anomalies and / or deviations in the workpiece-specific data set and / or in the workpiece support-specific data set, the size of the workpiece-specific data set and / or the size of the workpiece support-specific data set can be reduced.

[0087] Preferably, anomalies and / or deviations within the process steps of the manufacturing process are identifiable as quality defect causes using the process error data.

[0088] Anomalies and / or deviations within process steps of the manufacturing process can be associated with systematic quality defects classified by the cluster method using the process error data, in particular using analytical methods, preferably automatically.

[0089] In the form of a method, the method is carried out in a process technical installation, in particular in a painting installation, which has a number of different processing stations in which one or more process steps of a production process, in particular of a painting process, can be carried out.

[0090] Preferably, one or more particularly quality-related process data are detected within each processing station for each process step of the manufacturing process.

[0091] In the embodiment of the method, the workpiece-specific data set and / or the workpiece support-specific data set are continuously or discontinuously supplemented with process data, in particular quality-relevant.

[0092] In the method embodiment, the workpiece-specific data set and / or the workpiece support-specific data set are each supplemented with, in particular, quality-related process data, which each have a time stamp by means of which the respective process data can be uniquely associated with a point in time and / or a process step of the manufacturing process.

[0093] In the method aspect, the supplemented workpiece-specific data set of the workpiece and / or the supplemented workpiece support-specific data set of the workpiece support are stored in a database while and / or after the workpiece and / or workpiece support pass through the manufacturing process.

[0094] The invention further relates to a quality analysis system for analyzing quality defects of workpieces, preferably vehicle bodies and / or vehicle attachment parts, in particular after and / or during the passage through a manufacturing process in a method technical facility, in particular after and / or during the passage through a painting process in a painting facility.

[0095] The object of the present invention is to provide a quality analysis system for analyzing quality defects of workpieces, preferably vehicle bodies and / or vehicle attachment parts, after and / or during a manufacturing process in a method-technical installation, in particular after and / or during a painting process in a painting installation, in which quality defects can be avoided and / or causes of quality defects in the manufacturing process can be determined, avoided and / or eliminated.

[0096] This problem is solved according to the invention by a quality analysis system for analyzing quality defects of workpieces, preferably vehicle bodies and / or vehicle attachment parts, in particular after and / or during passage through a production process in a method-technical installation, preferably after and / or during passage through a painting process in a painting installation, having the features of claim 15.

[0097] The quality analysis system preferably comprises: - an interface for communication with a control system of a method-technical installation, in particular a painting installation; a control device, which is arranged and formed in such a way that the method according to any one of claims 1 to 14 can be carried out by means of the control device.

[0098] The invention further relates to a process technical installation, in particular a painting installation, comprising: - A control system capable of controlling the manufacturing process, in particular the painting process; A quality analysis system according to the present invention.

[0099] The process technical installation preferably comprises several treatment stations, in particular several painting stations.

[0100] For example, a painting installation may have a painting line with a number of process stations connected to each other.

[0101] Alternatively, or in addition, it is conceivable that the painting installation comprises a number of processing boxes, where each processing box comprises one or more processing stations.

[0102] The coating installation preferably comprises one or more of the following processing stations: - Pre-treatment station; -Cathode dip painting station; - Dryer after the cathode dip coating station; - 1st cabin; -1st dryer; - Base coat cabin; -Base coat dryer; -Clear coat cabin; -Clear coat dryer.

[0103] Further characteristics and / or advantages of the invention are the subject of the following description of an embodiment and the accompanying drawings. [Brief description of the drawings]

[0104] [Figure 1] FIG. 1 is a schematic representation of a process technical installation, in particular a painting installation. [Diagram 2] FIG. 2 is a graphical representation of an embodiment of a workpiece-specific data set for multiple workpieces. [Diagram 3] FIG. 3 illustrates diagrammatically a cluster method for classifying the quality data of the workpiece-specific data set shown in FIG. [Figure 4] FIG. 4 diagrammatically illustrates another embodiment of a workpiece specific data set for multiple workpieces. [Diagram 5] FIG. 5 illustrates diagrammatically a cluster method for classifying the quality data of the workpiece-specific data set shown in FIG. [Figure 6] FIG. 6 diagrammatically illustrates another embodiment of a workpiece specific data set for multiple workpieces. [Figure 7] FIG. 7 illustrates diagrammatically a cluster method for classifying the quality data of the workpiece-specific data set shown in FIG. [Figure 8] FIG. 8 shows a schematic visualization of systematic quality defects identified using the cluster method. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0105] In all figures, identical or functionally equivalent parts are provided with the same reference numbers.

[0106] FIG. 1 shows a process technical installation, generally designated 100 .

[0107] The process technical plant 100 is in particular a painting plant 102 .

[0108] The process technical installation 100 preferably comprises a number of processing stations 104 .

[0109] Here, a process technical installation 100, in particular a painting installation 102, comprises seven treatment stations 104.

[0110] Preferably, with the method technology installation 100 a manufacturing process can be carried out on a workpiece 106, in particular on a vehicle body 108 and / or on a vehicle attachment part.

[0111] In particular, within the processing stations 104, one or more respective manufacturing processes may be carried out.

[0112] The process technical installation 100, in particular the painting installation 102, shown in FIG.

[0113] Within the coating line 110, the processing stations 104 are preferably connected to one another.

[0114] For example, the workpiece 106 and / or workpiece support passes through the following processing stations 104 in the order listed.

[0115] In that case, it is conceivable, for example, that while the workpieces 106 pass through the processing station 104, only an individual workpiece 106, for example a vehicle body 108, is positioned on a workpiece support, which is not shown in the drawing.

[0116] Alternatively, or in addition, it is possible that multiple workpieces 106, for example a vehicle body not shown in the drawing, are arranged together on a single workpiece support, and the workpiece support carrying the workpieces 106 passes through the processing station 104.

[0117] A workpiece 106 , such as a vehicle body 108 , is pre-treated in a pre-treatment station 112 and transferred from the pre-treatment station 112 to a cathodic dip coating station 114 .

[0118] From the cathodic dip coating station 114, the workpiece 106, after having a coating provided thereon, is transferred after the cathodic dip coating station 114 to a dryer 116.

[0119] After the coating applied onto the workpiece 106 in the cathodic dip coating station 124 is dried in the dryer 116, the workpiece 106 is preferably transported into a basecoat cabin 118 where a coating is again applied onto the workpiece 106.

[0120] After the coating is applied in the basecoat cabin 118 , the workpiece 106 is preferably transferred into a basecoat dryer 120 .

[0121] After the coating applied onto the workpiece 106 in the basecoat cabin 118 is dried in a basecoat dryer 120, the workpiece 106 is preferably transported into a clearcoat cabin 122 where another coating is applied onto the workpiece 106.

[0122] After the coating is applied in the clearcoat cabin 122 , the workpiece 106 is preferably fed to a clearcoat dryer 124 .

[0123] After the coating applied to the workpiece 106 in the clearcoat cabin 122 is dried in a clearcoat dryer 124, the workpiece 106 is preferably fed to a control station 126 at the end of the manufacturing process.

[0124] Within the control station 126, quality control is preferably carried out by a quality inspector, for example using visual control.

[0125] The process technology installation 100 , in particular the painting installation 102 , preferably further comprises a control system 128 by means of which the manufacturing process, in particular the painting process, can be controlled in the treatment station 104 .

[0126] Furthermore, it may be advantageous if the process technical plant 100, in particular the painting plant 102, comprises a quality analysis system 130.

[0127] The quality analysis system 130 preferably comprises an interface 132 for communicating with the control system 128 of the process technical installation 100 , in particular the painting installation 102 .

[0128] The quality analysis system 130 is preferably configured to analyze quality defects of the workpiece 106, in particular the vehicle body 108 and / or vehicle attachments.

[0129] Preferably, using the control system 128, and preferably using the quality analysis system 130, a workpiece-specific data set 134 that is uniquely associated with the workpiece 106 is formed at the start of the manufacturing process, in particular at the start of the painting process.

[0130] The workpiece-specific data sets 134 for the workpieces 106 preferably each form a “digital workpiece” and / or a “digital image” of the respective workpiece 106 .

[0131] At the start of the manufacturing process, in particular at the start of the painting process, a workpiece-specific data set 134 is preferably formed automatically, for example by reading in the application data, for example using an RFID chip (not shown) arranged on the workpiece 106.

[0132] The workpiece-specific data set 134 therefore preferably comprises a workpiece-specific application data set 138, such as a unique workpiece identification number, a model type of the respective workpiece 106 and / or a color code for the color to be applied onto the workpiece 106 in the painting process.

[0133] Additionally, it may be advantageous for the workpiece-specific data set 134 to be supplemented with process data 140, particularly quality-related data, as the workpiece 106 moves through a manufacturing process, particularly a painting process.

[0134] Each workpiece-specific data set 134 is preferably stored in a database 142. In particular, the supplemental workpiece-specific data set 134 for the workpiece 106 is stored in the database 142 while and / or after the workpiece 106 moves through the manufacturing process.

[0135] The workpiece-specific data set 134 supplements the process data 140 , preferably further comprising a manufacturing data set 144 that includes the process data 140 .

[0136] It may be advantageous if the workpiece-specific data set 134 is supplemented with particularly quality-related process data 140 at each individual process step of the manufacturing process.

[0137] The process data 140, in particular the quality-related process data 140 that supplements the respective workpiece-specific data set 134 of the workpiece 106, preferably each have a time stamp, by means of which the respective process data can be uniquely associated with a point in time and / or a process step of the manufacturing process.

[0138] In particular, the quality-related process data 140 preferably comprises physical and / or chemical influence parameters for each coating and / or paint layer that is applied to the workpiece 106 in the manufacturing process.

[0139] The workpiece-specific data set 134 of each workpiece 106 is preferably supplemented with quality data 146, which quality data comprises information about quality defects of the respective workpiece 106, in particular about paint defects.

[0140] In particular, the workpiece-specific data set 134 can be supplemented with quality data 146 at the end of the manufacturing process, in particular at the end of the painting process.

[0141] The quality data 146 are preferably determined within the framework of a quality control by a quality inspector in the control station 126, for example by means of a visual control.

[0142] For example, the quality data 146 may include information regarding the location, type, size, and / or number of quality defects on each workpiece 106 .

[0143] In this case, quality defects, in particular coating defects, include, for example, coating errors, in particular dirt inclusions, holes, runners and / or air bubbles (boiling), different layer thicknesses, undesirable color shifts and / or deviations from predetermined process result parameters.

[0144] Processing result parameters include, for example: the thickness of the coating; the flatness of the coating; the uniformity of the layer thickness of the coating; the color and / or brightness of the coating; the hardness of the coating; the chemical composition of the coating; in particular the degree of crosslinking and / or the solvent content; and / or the degree of contamination of the coating.

[0145] The quality data 146 , which supplements the workpiece-specific data set 134 for each workpiece 106 , preferably further comprises information regarding the target status of each workpiece 106 .

[0146] In that case, the quality data 146 preferably includes information as to whether the workpiece 106 must be post-processed, what post-processing must be performed, whether post-processing is not necessary and / or whether post-processing is not possible.

[0147] As post-processing, for example one or more of the following post-processing steps are conceivable: polishing, spot repair and / or a new run through of the manufacturing process, in particular a painting process.

[0148] As process data 140, in particular as quality-related process data 140, which supplement the workpiece-specific data set 134, preferably one or more of the following process parameters are used: - Exceeding target times for process steps in the manufacturing process; -Events that occur while moving through process steps in the manufacturing process; -Body-in-white quality data for each workpiece; -Anomalies and / or deviations in the manufacturing process; -Weather data; -People data about people involved in a process step as they move through the process step in the manufacturing process.

[0149] The quality data of the body in white in particular comprises information regarding the sheet quality of the workpiece 106 or parts of the workpiece 106, such as the bonnet, the doors, the side parts and / or the roof.

[0150] To supplement the respective workpiece-specific data set 134, one or more of the following process parameters may be advantageously used: - the nozzle temperature in the dryers 116, 120, 124; -Paint amount; -Paint colors; -Paint charge; -Robot lane trajectory - cabin temperature in the painting cabins 118, 122; - cabin humidity in the painting cabins 118, 122; - the bath temperature and filling state in the cathode dip coating station 114 and in the pretreatment station 112; -Transport movement of the workpiece through the cathodic dip coating station 114; - the concentration of chemicals in the cathode dip coating station 114; - the voltage in the cathode dip coating station 114; - Temperatures and run-through times in the dryers 116, 120, 124.

[0151] The process parameters used to supplement each workpiece-specific data set 134 are preferably pre-processed.

[0152] It is conceivable, in particular, that process parameters with a high variability in time, such as mechanical, hydraulic or electrical process parameters, are preprocessed, but the preprocessing of process parameters with a low variability in time, i.e. inertial process parameters, such as thermodynamic process parameters, can preferably be omitted.

[0153] The events comprise, for example, maintenance events, alarms from the control system 128 of the process technical installation 100, information about shift changes, information about production stoppages, information about entries into the processing stations 104 of the process technical installation 100, information about cleaning interruptions.

[0154] Maintenance events added to the workpiece-specific data set 134 can be used, for example, to identify systematic quality defects that occur after maintenance.

[0155] Preferably, furthermore, one or more of the following process parameters are used to supplement the respective workpiece-specific data set 134: -Transfer technical information; - the residence time of the workpiece 106 in the manufacturing process; "Overnight storage" of the workpiece 106 inside the process technical installation 100, for example inside the dryers 116, 120, 124.

[0156] For example, the residence time of the workpiece 106 in the dryers 116, 120, 124 is used to supplement the respective workpiece-specific data set 134. That is, quality defects can be inferred from exceeding a target time for the presence of the workpiece 106 in the dryers 116, 120, 124 in particular.

[0157] The anomalies are preferably determined automatically from the deviation between the actual process values ​​and the target process window within a process step of the manufacturing process. 4 The temperature is determined using one or more sensors (not shown).

[0158] Alternatively or additionally, it is conceivable that the anomalies are automatically determined from deviations between actual process values ​​of process steps of the manufacturing process and normal states learned using machine learning methods for each process step.

[0159] Preferably, the workpiece-specific data set 134 is supplemented with process error data 148 in the event of anomalies and / or deviations within the process steps of the manufacturing process, particularly while the workpiece 106 passes through each process step of the manufacturing process.

[0160] The process error data 148 preferably comprises information regarding anomalies and / or deviations within the process steps of a manufacturing process.

[0161] The process error data 148 is specifically an “error marker.” For example, the process error data 148 can be used to mark the presence or absence of anomalies and / or deviations in each process step as the workpiece 106 moves through the process steps.

[0162] Using the process error data 148, anomalies and / or deviations within each process step of the manufacturing process can preferably be identified as errors.

[0163] By using the process error data 148 to identify anomalies and / or deviations within the process data 140 of the workpiece-specific data set 134, the size of the workpiece-specific data set 134 may be preferably reduced.

[0164] FIG. 2 illustrates an embodiment of the workpiece-specific data sets 134 for various workpieces 106 .

[0165] In a first column, for example, application data 136 of the respective workpiece-specific data set 134 are shown, for example in the form of a color code and / or in the form of a workpiece identification number.

[0166] Columns 2 to 7 each show process data 140 of a respective workpiece-specific data set 134 from a process step of the manufacturing process in the method technology installation 100, in particular in the processing station 104 of the painting installation 102.

[0167] Exclamation points are used to specifically indicate process error data 148, or "error markers," that contain information regarding anomalies and / or deviations within a process step of a manufacturing process.

[0168] In column 9 further quality data 146 is shown, which preferably comprises information regarding the location, type, size and / or number of quality defects.

[0169] For example, for workpiece W1, anomalies and / or deviations in pre-treatment station 112 are stored as process error data 148.

[0170] Workpieces W3 through W5 have been identified for anomalies and / or deviations, for example, in the dryer 116 after the cathodic dip coating station 114, and characterized with process error data 148.

[0171] Workpiece W8 is further characterized using process error data 148, for example, for anomalies and / or deviations within basecoat dryer 120.

[0172] Quality data 146, which contains information about quality defects recognized in the control station 126, is stored, for example, in the ninth column. For example, the quality data 146 here contains information about the type of quality defect.

[0173] Preferably, the quality data 146 of the multiple workpiece-specific data sets 134 are automatically classified into different quality defect clusters 150 using the quality analysis system 130 using cluster methods to recognize systematic quality defects.

[0174] The quality data 146 of the workpiece-specific data set 134 is classified in a cluster method, preferably using one or more classification rules.

[0175] In that case, it may be advantageous if the quality data 146 of the workpiece-specific data set 134 is classified into systematic quality defect clusters 150 and non-systematic quality defect clusters.

[0176] To classify the quality data 146, preferably expert defined and / or learned classification rules are used.

[0177] The classification rules defined by the experts may, for example, have within a moving window a defined number of workpieces 106 that pass through the manufacturing process and have a defined number of workpieces 106 that have identical and / or similar quality defects.

[0178] Here, when identical and / or comparable quality defects are recognized in three or more workpieces 106 among five workpieces 106 that are consecutive to one another in the manufacturing process, the quality data 146 of the workpiece-specific data set 134 is classified into a systematic quality defect cluster 150.

[0179] The learned classification rules are preferably learned from the quality data 146 of the workpiece-specific data set 134 using machine learning methods.

[0180] The classification rules can then be learned using supervised and / or unsupervised machine learning methods.

[0181] Preferably, the classification of the quality data 146 is performed continuously using the cluster method. Alternatively, or in addition, the classification of the quality data 146 can be performed in a batch process using the cluster method.

[0182] Preferably, the systematic quality defects identified using the cluster method are displayed to the user using the visualization 152 of the quality analysis system 130 shown in FIG. 8, for example as a real-time report in an alarm system and / or as an analysis report in an analysis system.

[0183] In the visualization 152 shown in FIG. 8, all quality defects of the quality defect cluster 150 are located in the workpiece 106 , in particular in the left side door of the vehicle body 108 .

[0184] FIG. 3 illustrates the implementation of the cluster method for the workpiece-specific data set 134 shown in FIG.

[0185] At time points T1 and T2, the quality analysis system 130 has not yet recognized the quality defect cluster 150 having systematic quality defects.

[0186] At time T3, the quality analysis system 130 is used to recognize a quality defect cluster 150 because the quality data 146 of the workpiece-specific data set 134 for workpieces W3, W4, and W7 each have the same quality defect F1.

[0187] For example, it is contemplated that the workpiece 106 may be rejected from the manufacturing process based on the quality defect clusters 150 identified using the cluster methodology. Alternatively, or in addition, the process control of the manufacturing process may be adapted using the control system 128 based on the quality defect clusters 150 identified by the cluster methodology.

[0188] Preferably, quality defect causes are determined automatically using the analysis method for systematic quality defects that have been recognized using the cluster method and classified into various quality defect clusters 150. Preferably, in this case, in particular the quality-related process data 140 of the workpiece-specific data set 134 of the workpieces W3, W4 and W7 having systematic quality defects of the quality defect clusters 150 are analyzed.

[0189] In the analysis method, the quality defect causes are preferably determined using one or more analysis rules.

[0190] To determine the quality defect causes, preferably expert defined and / or learned analysis rules are used.

[0191] The analysis rules to be trained are preferably trained using machine learning methods, for example, where it is conceivable that the analysis rules are trained using supervised and / or unsupervised machine learning methods.

[0192] Anomalies and / or deviations within the process steps of the manufacturing process are preferably identifiable as quality defect causes using the process error data 148 .

[0193] Anomalies and / or deviations within the process steps of the manufacturing process can be automatically associated with systematic quality defects classified using clustering methods, preferably using the process error data 148, in particular using analytical methods.

[0194] Preferably, analytical methods are used to determine the relationships between the identified quality defect causes and the systematic quality defects classified into the various quality defect clusters 150 .

[0195] For example, it is conceivable that an analysis rule defined by an expert may be used that forms a relationship between an anomaly in the dryer 116 behind the cathode dip coating station 114 and the quality defect F1 of the quality defect cluster 150.

[0196] Preferably within the visualization 152 of the quality analysis system 130, within a window 154 of the visualization 152, the quality defect causes for the quality defect cluster 150 shown in FIG.

[0197] The defined quality defect causes preferably comprise anomalies and / or deviations within the manufacturing process, in particular within the painting process.

[0198] Using the determined relationship, impending systematic quality defects are preferably automatically determined by inferring future quality defects of the workpiece 106 from one or more anomalies and / or defects identified in the ongoing manufacturing process using the analytical methodology using the determined relationship as the workpiece 106 moves through the manufacturing process.

[0199] In that case, the workpiece 106 is preferably associated with quality defect clusters 150 already as it goes through the manufacturing process using analysis methods.

[0200] Preferably, therefore, systematic quality defects which cannot be recognized within the framework of the quality control in the control station 126 can also be recognized by a posteriori inference from the manufacturing process.

[0201] The embodiment of the workpiece-specific data set 134 for the various workpieces 106 shown in FIG. 4 is substantially distinguished from the embodiment of the workpiece-specific data set 134 for the various workpieces 106 shown in FIG. 2 in that the quality data 146 of the workpiece-specific data set 134 for workpieces W6 to W8 each contain information regarding quality defect F2.

[0202] At time T4, a quality defect cluster 150 is recognized while applying the classification rules using the cluster method because the quality data 146 of the workpiece-specific data sets 134 of workpieces W6 to W8 each have the same quality defect F2 (see FIG. 5).

[0203] The process data 140 of the workpiece specific data set 134 includes information about the cabin temperature in the basecoat cabin 118 in a fifth column.

[0204] The cabin temperature has already started to rise at workpieces W4 and W5, and it is then possible that the cabin temperature in base coat cabin 118 is outside the target process window defined for the cabin temperature in base coat cabin 118 at workpieces W6 to W8, respectively.

[0205] Using analytical methods, the cabin temperature within the basecoat cabin 118 is preferably identified as a quality defect cause.

[0206] For example, when implementing the analysis method, an analysis rule defined by an expert is used, which forms a relationship between the cabin temperature in the base coat cabin 118 and the quality defect F2 of the quality defect cluster 150.

[0207] The implementation of the cluster method and the analysis method in the embodiment of the workpiece-specific dataset 134 shown in Figures 4 and 5 is substantially identical to the implementation of the cluster method and the analysis method in the embodiment of the workpiece-specific dataset 134 shown in Figures 2 and 3, so reference is made to the above description thereof to that extent.

[0208] The embodiment of the workpiece-specific data 134 for the various workpieces 106 shown in FIG. 6 is substantially distinguished from the embodiment of the workpiece-specific data set 134 for the various workpieces 106 shown in FIG. 2 in that the quality data 1456 for workpieces W2 and W3 each contain information regarding quality defect F1, and the quality data 146 for workpieces W4 to W7 each contain information regarding quality defect F2.

[0209] The quality defects F1 and F2 are, for example, identical and / or similar and have already been recognized as a quality defect cluster 150 at time T1 by applying classification rules using the cluster method (see FIG. 7).

[0210] The process data 140 of the workpiece specific data set 134 includes, in a fourth column, information regarding target time exceedance in the dryer 116 after the cathodic dip coating station 114 .

[0211] This is especially true for workpieces W2 and W3 having a quality defect F1.

[0212] Of course, workpieces W4 through W7 following workpieces W2 and W3 also meet the target time overrun in dryer 116 after cathode dip coating station 114.

[0213] Using analytical methods, exceeding the target temperature in the dryer 116 after the cathode dip coat station 114 is identified as a quality defect cause, preferably for quality defects F1 and F2, and in particular for quality defect cluster 150.

[0214] The implementation of the cluster method and the analysis method in the embodiment of the workpiece-specific dataset 134 shown in Figures 6 and 7 is substantially identical to the implementation of the cluster method and the analysis method in the embodiment of the workpiece-specific dataset 134 shown in Figures 2 and 3, so reference is made to the above description thereof to that extent.

[0215] The method steps described with reference to FIGS. 1 to 8 can not only be performed for an individual workpiece 106 using the quality analysis system 130 .

[0216] In particular, it is conceivable that the method steps are performed using the quality analysis system 130 for a plurality of workpieces 106 each commonly arranged on a workpiece support, e.g. also for a plurality of vehicle mounting parts not shown in the drawing, while the workpiece support and the workpieces 106 arranged thereon pass through the processing station 104.

[0217] In that case, preferably, instead of the workpiece-specific data set 134, a workpiece support-specific data set is used.

[0218] The workpiece support-specific data set is in particular uniquely associated with the workpiece support and with all workpieces 106 respectively arranged on the workpiece support, for example with the workpiece support and with all vehicle attachments arranged thereon.

[0219] In addition, the method steps that can be implemented using the quality analysis system 130 correspond to the method steps described above, and reference is made to the description thereof in this regard.

[0220] Overall, a method for analyzing quality defects of a workpiece 106, preferably a vehicle body 108 and / or a vehicle attachment part, after or when passing through a manufacturing process in the process technical installation 100, preferably after or when passing through a painting process in the painting installation 102, can be provided, with which quality defects can be avoided and / or with which quality defect causes in the manufacturing process can be determined, avoided and / or eliminated. Furthermore, a quality analysis system 130 can be provided for implementing such a method for analyzing quality defects of the workpiece 106.

Claims

1. A method for analyzing a vehicle body (108) and / or vehicle attachment parts as a workpiece for quality defects after and / or while passing through a painting process in a painting facility (102), the method comprising: - forming a workpiece-specific data set (134) uniquely associated with the vehicle body (108) or the vehicle attachment at the start of the painting process and / or forming a workpiece support-specific data set uniquely associated with the workpiece support at the start of the painting process; - supplementing a workpiece-specific data set (134) with quality-related process data (140) while the vehicle body (108) or the vehicle attachment parts go through a painting process and / or supplementing a workpiece support-specific data set with quality-related process data while the workpiece support goes through a painting process; storing a workpiece-specific data set (134) in a database (142) and / or storing a workpiece support-specific data set in a database (142); Including, the workpiece-specific data set (134) and / or the workpiece support-specific data set are supplemented at the end of the painting process by quality data (146), said quality data comprising information about quality defects of the respective vehicle body (108) or vehicle attachment part, The quality data (146) of the plurality of workpiece-specific data sets (134) and / or the plurality of workpiece support-specific data sets are automatically classified into various quality defect clusters (150) using a cluster method in order to recognize systematic quality defects; expert-defined and / or learned classification rules are used to classify the quality data (146); To supplement the respective workpiece-specific data set and / or the respective workpiece support-specific data set, the following process parameters are determined: - nozzle temperature in the dryer; -Amount of paint; -Paint color; - paint charge; - Robot lane trajectory; - cabin temperature in the painting cabin; - cabin humidity in the painting cabin; - the bath temperature and filling state in the cathodic dip coating station and in the pretreatment station; - Transport movement of the vehicle body (108) or vehicle attachment parts through the cathodic dip painting station; - Chemical concentrations in the cathodic dip coating station; - voltage in the cathodic dip coating station; - temperature and time of passage in the dryer, The method of claim 1, wherein one or more of the following is used:

2. 2. The method according to claim 1, characterized in that the quality defect causes for the systematic quality defects recognized using the cluster method and classified into various quality defect clusters (150) are automatically determined using the analysis method by analyzing the quality-related process data (140) of the workpiece-specific dataset (134) of the vehicle body (108) and / or vehicle attachment parts having systematic quality defects of the respective quality defect cluster (150) and / or by analyzing the quality-related process data (140) of the workpiece support-specific dataset of the workpiece support having the vehicle body (108) and / or vehicle attachment parts having systematic quality defects of the respective quality defect cluster (150).

3. 3. The method according to claim 2, characterized in that the determined quality defect causes comprise anomalies and / or deviations within the painting process.

4. 4. The method according to claim 2 or 3, characterized in that an analytical method is used to determine the relationship between the determined quality defect causes and the systematic quality defects classified into different quality defect clusters (150).

5. 5. The method according to claim 4, characterized in that future quality defects of the vehicle body (108) or vehicle attachment parts are automatically estimated by the determined relationship using an analytical method from one or more determined anomalies and / or deviations in the ongoing painting process, thereby automatically determining impending systematic quality defects while the vehicle body (108) or vehicle attachment parts are passing through the painting process using the determined relationship.

6. 6. The method according to claim 1, wherein the workpiece-specific data set (134) and / or the workpiece support-specific data set are supplemented within each individual process step by quality-related process data (140).

7. As quality-related process data (140) that complement the workpiece-specific data set (134) and / or the workpiece support-specific data set, the following process parameters are included: - Exceeding target times within process steps of the painting process; - Events that occur during the passage through the process steps of the painting process; - quality data of the body-in-white of each vehicle body (108) or vehicle attachment; - anomalies and / or deviations within the painting process; - weather data; - people data about people involved in the process steps as they move through the process steps of the painting process; 7. The method according to claim 1, wherein one or more of the following are used:

8. A method according to any one of claims 1 to 7, characterized in that while the vehicle body (108) or vehicle mounting parts and / or the workpiece support passes through a process step of the painting process, the workpiece-specific data set (134) and / or the workpiece support-specific data set are supplemented with process error data (148), which process error data comprises information on anomalies and / or deviations within the process steps of the painting process.

9. 9. The method according to claim 1, characterized in that the method is carried out in a painting installation (102) having a plurality of mutually separated treatment stations (104), in which one or more process steps of the painting process can each be carried out.

10. 10. The method according to claim 1, wherein the workpiece-specific data set (134) and / or the workpiece support-specific data set (140) are supplemented, continuously or discontinuously, with quality-related process data (140).

11. 11. The method according to claim 1, wherein the workpiece-specific data set (134) and / or the workpiece support-specific data set are each supplemented with quality-related process data (140), the process data each having a time stamp, by means of which the respective process data (140) can be unambiguously associated with a time point and / or a process step of the painting process.

12. A method according to any one of claims 1 to 11, characterized in that the supplemented workpiece-specific data set (134) of the vehicle body (108) or vehicle mounting parts and / or the workpiece support and / or the supplemented workpiece support-specific data set of the workpiece support are stored in a database (142) while and / or after the vehicle body (108) or vehicle mounting parts and / or the workpiece support passes through a painting process.

13. A quality analysis system (130) for analyzing a vehicle body (108) and / or vehicle attachment parts for quality defects after and / or while passing through a painting process in a painting facility (102), comprising: - having an interface (132) for communicating with the control system (128) of the painting installation (102); - a control device, said control device being arranged and configured in such a way that the method according to any one of claims 1 to 12 can be carried out with said control device, Quality analysis system.

14. A painting system (102), comprising: - having one or more processing stations (104); - a control system (128) by means of which the painting process can be controlled in one or more treatment stations (104); - a quality analysis system (130) according to claim 13, Painting equipment.

Images