Computer-implemented method for detecting spatially related manufacturing anomalies in the production of multiple motor vehicles

Abstract

The invention relates to a computer-implemented method for detecting spatially related manufacturing anomalies in the production of multiple motor vehicles, comprising the following steps: - Detection of manufacturing anomalies in the production of motor vehicles, whereby position data is recorded for each detected manufacturing anomaly; - Determining the spatial proximity of manufacturing anomalies using positional data; and - Grouping of the manufacturing anomalies into distinguishable groups (2; 3; 4; 5) depending on the determined spatial proximity.

Description

[0001] The present invention relates to a computer-implemented method for detecting spatially related manufacturing anomalies in the production of multiple motor vehicles according to claim 1.

[0002] It is known from the prior art to automatically detect manufacturing anomalies. It is also known to automatically determine their spatial position. US Patent 2010 / 0241380A1 discloses a method for analyzing damage to a motor vehicle. This method also determines damage within a spatial area and the frequency with which it occurs.

[0003] In contrast, the present invention aims to better identify the causes of manufacturing anomalies.

[0004] This problem is solved by a method according to claim 1 and by a system according to claim 9. Embodiments of the invention are specified in the dependent claims.

[0005] The method according to claim 1 comprises detecting manufacturing anomalies during the production of motor vehicles. The detection can be performed, for example, automatically and / or manually. Positional data is recorded for each detected manufacturing anomaly. This positional data can preferably be three-dimensional. The positional data can, for example, contain information about the position of the respective manufacturing anomaly on or in the respective motor vehicle.

[0006] The spatial proximity of manufacturing anomalies is determined using positional data. The manufacturing anomalies are then grouped into distinguishable groups based on this determined spatial proximity. Metadata is also used in this grouping process. Therefore, it is possible that the grouping is not based solely on spatial proximity. Metadata is recorded for each manufacturing anomaly. This metadata includes information about a production step at or immediately before the detection of the manufacturing anomaly. Preferably, the grouping is performed independently of any previously defined vehicle areas.

[0007] Grouping defects based on spatial proximity is advantageous for identifying systematically occurring manufacturing anomalies. Since manufacturing anomalies detected in different vehicles are grouped together, the large number of defects allows for the identification of patterns or dependencies. This can reveal connections in the development of the manufacturing anomalies that might otherwise only be detectable with significant analytical effort, or not at all. This saves time in conducting the analysis. One or more causes can be eliminated more quickly. Manufacturing quality can therefore be improved.

[0008] Spatial areas are defined during grouping, and not before grouping, in which the manufacturing anomalies of each group are positioned. These areas can also overlap, for example.

[0009] The metadata is used for grouping. Preferably, the grouping is performed based on the metadata. Using metadata for grouping can be particularly advantageous for identifying patterns or dependencies in the development of manufacturing anomalies. The mere spatial proximity of manufacturing anomalies does not necessarily imply that they are related. However, if the metadata also shows a high degree of similarity, this provides further evidence of a relationship between the manufacturing anomalies.

[0010] According to one embodiment of the invention, the manufacturing defects can be visually displayed as groups. For example, the groups can be distinguished from one another by color. This makes it easier for a user to differentiate the groups.

[0011] According to one embodiment of the invention, the manufacturing defects can be displayed three-dimensionally in the optical output. For example, the manufacturing defects can be displayed on and / or within a three-dimensional model of a motor vehicle. This makes it even easier for a user to distinguish the groups from one another. It is also possible for the user to be presented with a suggestion for a grouping of the manufacturing defects deemed meaningful by an algorithm or artificial intelligence. This suggestion can, for example, be displayed as a message or notification on a screen. Alternatively, the suggested grouping can be displayed on and / or within the three-dimensional model itself.

[0012] According to one embodiment of the invention, the metadata can include information about the time of detection of the respective manufacturing anomaly, about an anomaly category of the respective manufacturing anomaly and / or about a motor vehicle in which the respective manufacturing anomaly was detected.

[0013] In this context, the term "anomaly category" can be understood to refer specifically to a type of manufacturing anomaly. For example, there could be a first anomaly category for manufacturing anomalies affecting the paintwork of a particular vehicle, and a second anomaly category for manufacturing anomalies affecting sheet metal parts of the same vehicle. Paint damage, for instance, could be assigned to the first anomaly category. Sheet metal deformation, for example, could be assigned to the second anomaly category.

[0014] The information about the motor vehicle may, for example, relate to a model of the motor vehicle or to a specific individual motor vehicle.

[0015] According to one embodiment of the invention, the manufacturing anomalies can be displayed in the optical output depending on or together with the metadata. This is particularly advantageous if the metadata includes information about the time of detection of the respective manufacturing anomaly. In this case, by displaying the manufacturing anomalies depending on the respective time, correlations and regularities can be more easily identified.

[0016] According to one embodiment of the invention, grouping parameters can be received via user input. These grouping parameters can then be used during the grouping process. For example, the grouping parameters can define how the metadata is used during grouping. Specifically, the grouping parameters can define how similar the information contained in the metadata must be to each other for the respective manufacturing anomalies to be grouped into the same group, preferably taking spatial proximity into account. In particular, with regard to information about the detection time, the grouping parameters can, for example, include a threshold value that must be undercut for the respective manufacturing anomalies to be grouped into the same group.

[0017] According to one embodiment of the invention, manufacturing anomalies can be detected in the production of different motor vehicles.

[0018] According to one embodiment of the invention, some manufacturing anomalies can be excluded from any group depending on their determined spatial proximity. For example, if a required spatial proximity is defined, it is possible to exclude a specific manufacturing anomaly that does not exhibit this required spatial proximity to any of the other manufacturing anomalies from any group. A similar approach can be implemented if a required temporal proximity is defined. In this case, it is possible to exclude a specific manufacturing anomaly that does not exhibit this required temporal proximity to any of the other manufacturing anomalies from any group. It is also possible to define both the required spatial and the required temporal proximity.

[0019] The system according to claim 9 comprises a digital data storage device and a processing unit. The processing unit can, for example, also be referred to as a processor. Instructions are stored in the data storage device, which can be read and executed by the processing unit. The instructions are configured to cause the processing unit, upon execution of the instructions, to carry out a method according to an embodiment of the invention.

[0020] Further features and advantages of the present invention will become clear with reference to the following description of preferred embodiments and the accompanying figures. The same reference numerals are used for identical or similar components, features, or elements, and for components, features, or elements with identical or similar functions. Fig. 1. A schematic perspective view of a 3D model of a motor vehicle with marked position data of detected manufacturing anomalies; and Fig. 2 a schematic representation of information about the times of detection of the manufacturing anomalies from Fig. 1.

[0021] The in Fig. The 3D model 1 of the motor vehicle shown is used to illustrate manufacturing anomalies detected during the production of multiple motor vehicles. This model could, for example, be a 3D model based on the motor vehicles that are being or have been manufactured. However, it is also possible to use a general 3D model based on any type of motor vehicle.

[0022] In the context of this description, a manufacturing anomaly is understood to mean in particular a defect or deformation of a component or damage or visual impairment of a paint coating.

[0023] The detected manufacturing anomalies are in Fig. 1. The manufacturing defects are represented as points. Some of the manufacturing defects are already grouped. The ungrouped manufacturing defects will not be grouped, or will be grouped at a later time. A first group 2 of the manufacturing defects is spatially arranged in a first area located near the lower right edge of the windshield. A second group 3 of the manufacturing defects is arranged in a second area located in a right-hand edge region of the 3D model 1 below the A-pillar. The first area overlaps with the second area. A third group 4 of the manufacturing defects is arranged in a third area located at the lower end of a rear wheel arch. A fourth group 5 of the manufacturing defects is arranged in a fourth area directly adjacent to the third area.

[0024] Manufacturing anomalies are assigned to groups 2, 3, 4, and 5 based on the position data and metadata recorded for each anomaly. This means that the manufacturing anomalies are not grouped solely based on position data. This is advantageous for identifying patterns and / or dependencies.

[0025] If only the position data were used for grouping, at least some of the manufacturing anomalies of the second group 3 would be grouped into the first group 2, since their distance to the manufacturing anomalies of the first group 2 is less than to most of the manufacturing anomalies of the second group 3.

[0026] In Fig. 2 are examples of metadata, including information about the times of detection of manufacturing anomalies. Fig. Figure 1 shows the time plotted in the transverse direction. The individual groups are shown in separate rows. The manufacturing anomalies of the respective groups 2, 3, 4, and 5 were detected at similar times in the manufacturing process. Therefore, it makes sense to assign manufacturing anomalies detected at similar times to the same group, even if the spatial distance is sufficiently small. In this way, by defining different grouping parameters, the manufacturing anomalies can be assigned to different groups. The grouping parameters can determine the necessary temporal similarity of the detection and the necessary spatial proximity, both of which must be met for the manufacturing anomalies to be grouped into the same group. The grouping parameters can, for example, be received via user input.

[0027] The manufacturing anomalies of the third group 4 and the fourth group 5, for example, are spatially relatively close to each other (see Fig. 1) However, the detection times of the respective manufacturing anomalies are further apart than a time threshold defined by the grouping parameters, so they are not grouped into the same group. The same applies analogously to the manufacturing anomalies of the first group 2 and the second group 3.

Claims

[1] Computer-implemented method for detecting spatially related manufacturing defects in the production of multiple motor vehicles, comprising the following steps: - Detection of manufacturing anomalies in the production of motor vehicles, whereby position data is recorded for each detected manufacturing anomaly; - Determining the spatial proximity of manufacturing anomalies using position data, whereby metadata is recorded for each manufacturing anomaly, the metadata containing information about a production step at or immediately before the detection of the respective manufacturing anomaly; and - Grouping of the manufacturing anomalies into distinguishable groups (2; 3; 4; 5) depending on the determined spatial proximity, whereby the metadata is used in the grouping, whereby spatial areas are only defined during the grouping in which the manufacturing anomalies of each of the groups (2; 3; 4; 5) are positioned. [2] Method according to claim 1, characterized by , that the manufacturing anomalies are visually identifiable as groups (2; 3; 4; 5). [3] Method according to the previous claim, characterized by , that the manufacturing anomalies are displayed three-dimensionally in the optical output. [4] Method according to any one of the preceding claims, characterized bythat the metadata includes information about the time of detection of the respective manufacturing anomaly, about an anomaly category of the respective manufacturing anomaly and / or about a motor vehicle in which the respective manufacturing anomaly was detected. [5] Method according to the previous claim, characterized by , that the manufacturing anomalies in the optical representation are displayed depending on or together with the metadata. [6] Method according to any one of the preceding claims, characterized by , that grouping parameters are received through user input, and the grouping parameters are used during grouping. [7] Method according to any of the preceding claims, characterized by that manufacturing anomalies are detected in the production of different motor vehicles. [8] Method according to any one of the preceding claims, characterized by, that some of the manufacturing anomalies, depending on the determined spatial proximity, are not grouped into any of the groups (2; 3; 4; 5). [9] System comprising a digital data storage device and a processing unit, wherein instructions are stored in the data storage device which can be read and executed by the processing unit, wherein the instructions are configured to cause the processing unit, upon execution of the instructions, to carry out a method according to one of the preceding claims.

Images