Method for operating a production system, computer program, and electronically readable data carrier

Abstract

The invention relates to a method for operating a production system (1) in which blanks (2) as semifinished product are processed with specific process parameters to form components (4), comprising the steps of: determining a sequence in which the blanks (2) are intended to be processed in the production system (1) (S1), determining a discontinuity between successive blanks (2) on the basis of at least one semifinished product property of each blank (2) in question and / or a parameter and / or an environmental state of a cutting system (S2), classifying the determined discontinuity on the basis of at least one property value and / or historical data (S3), and outputting an instruction signal for the production system (1) according to the classification of the determined discontinuity (S4). The invention further relates to a computer program, to a data carrier and to a production system (1).

Description

[0001] 23-3760 PIF

[0002] 1

[0003] Method for operating a manufacturing plant, computer program and electronically readable data carrier

[0004] The invention relates to a method for operating a production plant in which, in particular, a change in process parameters can be specified, wherein circuit boards are processed into components in the production plant as semi-finished products with specific process parameters, according to claim 1. Furthermore, the invention relates to a computer program and an electronically readable data carrier according to claims 13 and 14.

[0005] In the production of components, particularly body parts, in press shops, which may be part of a manufacturing plant, various process steps are carried out. First, flat sheet metal blanks (or blanks) are cut from a coil on a coil line. The cut blanks are then stacked, allowing for traceability, and stored temporarily before further processing. In a press line or press within the production plant, the cut blanks, which are semi-finished products and are typically formed from flat sheet metal, especially thin sheet metal, are deep-drawn. This is followed by further process steps, such as trimming and / or reshaping. These different process steps can be decoupled both temporally and spatially.For example, the circuit boards can be cut in a coil plant of a supplier.

[0006] The semi-finished products processed in the manufacturing plants or press shops can exhibit variations in their properties. For example, a sheet metal covering, a tribological property such as the amount of lubricant, a roughness, and / or an elastoplastic material property can vary. Depending on the extent of these variations, it may be necessary to adjust process parameters affecting the production process and, in particular, the manufacturing plant. 23-3760 PIF

[0007] 2 in order to achieve a required quality of the produced components or the required component quality.

[0008] Adjusting process parameters typically involves downtime of the production plant and is carried out by a plant operator based on their experience. Depending on how long it takes to find the right combination of process parameters, significant costs can arise due to the downtime. Therefore, process parameter adjustments often only occur when quality problems arise or have already arisen.

[0009] Typical process parameters for presses or presses can include, for example, draw cushion forces, settings of the draw aids, the positioning of the guides, the kinematics of a ram movement, and / or the application of additional lubricant. Furthermore, the setting of the straightening unit on the coil line can also be adjusted via process parameters (of the production line).

[0010] Currently, semi-finished product properties, as well as indirect and / or direct data on the quality of the produced components or associated intermediate products, and corresponding process parameters of the production plant, such as a coil line or a press, can be recorded and assigned to a circuit board and, consequently, to a component. This assignment is typically done via a serial number, which is applied to the surface of the material, for example, during the cutting process. This data can be stored in a database, with a data record available for each produced circuit board and / or component. During the manufacturing process, the circuit boards are divided into several stacks. These stacks are then stored, and individual stacks are delivered from storage to the press as needed.In automotive production, the storage period for circuit boards can range from a few minutes to several months.

[0011] If, during the processing of circuit boards in a press shop, a first stack originating from a first coil is followed, for example, by a second stack originating from a second coil, a discontinuous change in the properties of the circuit board or the semi-finished product properties can occur at precisely the time when the first or uppermost circuit board of the second stack is processed. 23-3760 PIF

[0012] 3. In the worst case, if this jump or discontinuity remains undetected, it may happen that the components resulting from the circuit boards of the second stack do not meet the corresponding quality requirements, and therefore, if rework is not possible for technical or economic reasons, must be declared as scrap.

[0013] The object of the present invention is to provide a method, a computer program, and a data carrier by which discontinuities between circuit boards or stacks of circuit boards can be reacted to particularly advantageously.

[0014] This problem is solved according to the invention by the subject matter of the independent claims. Advantageous embodiments and further developments of the invention are shown in the dependent claims, the description, and in the drawing.

[0015] A first aspect of the invention relates to a method for operating a production plant, in particular for specifying a process parameter change for the production plant in which circuit boards, especially from at least one stack, are processed into components. The circuit boards are processed as semi-finished products with specific process parameters, particularly those of the production plant.

[0016] The method according to the invention comprises the following steps:

[0017] The first step involves determining the sequence in which the circuit boards or stacks are to be processed in the production plant.

[0018] In a second step, a discontinuity between, in particular two, successive circuit boards (and / or stacks) is determined based on at least one semi-finished product property of the respective circuit board and / or at least one parameter, for example during the cutting of the circuit boards and / or an environmental condition, in particular of a cutting system.

[0019] In a third step, the identified discontinuity is classified based on a property value and / or historical data and / or, in particular, the expected component quality. 23-3760 PIF

[0020] 4

[0021] In a fourth step, an instruction signal is issued to the production plant, depending on the classification of the detected discontinuity. The production plant is then operated specifically according to this instruction signal.

[0022] In principle, the method can be applied to any reference quantity, for example the circuit boards or the stacks, but also to coils, individual parts or individual circuit boards and / or orders and the like.

[0023] The invention is based on the understanding that discontinuities can occur, for example, when changing coils between circuit boards, which can affect component quality during the manufacturing process in the production plant. In some cases, adjusting process parameters may be sufficient to achieve a desired component quality. In other cases, the discontinuity may not affect the quality of the produced components at all, or not to such an extent that a change in process parameters would be necessary.

[0024] According to current technology, there is no record of the conditions under which a discontinuity in the properties of printed circuit boards could become critical. A wide variety of causes can lead to discontinuities in the properties of printed circuit boards with respect to the processing sequence:

[0025] Discontinuity can occur when all stacks are cut from a single coil. A seamless transition between sheets originating from the same coil can only be ensured if the sheets are processed in the press in the reverse order of how they were produced in the cutting line. Typically, several stacks are generated when a coil is cut. To maintain the required sequence and avoid discontinuity, the necessary order of the stacks must be observed during processing in the press or production line. If the processing order is reversed, discontinuity in the properties or semi-finished product characteristics of the sheet can occur at the transition between two stacks during processing in the press.If the order of the blanks within the stack is reversed, or if the stack is processed in the wrong order, discontinuities in the properties of the blanks or the semi-finished product properties can occur during the transition between two stacks when processing in presses. For example, if a first coil is cut on a blank slitting machine, n stacks of blanks can be formed. Typically, there are k blanks on each of the 23-3760 PIFs.

[0026] 5

[0027] Stack. An exception is the last stack, which results from the first coil. This usually contains I circuit boards, where l <k. Erfolgt anschließend das Schneiden eines zweiten Coils, wird zunächst dieser letzte Platinenstapel aufgefüllt. Das heißt, auf diesem Stapel werden k-l Platinen abgelegt, die aus dem zweiten Coil gebildet wurden. Somit kann innerhalb dieses letzten Stapels beispielsweise eine Unstetigkeitsstelle der Eigenschaften der Platinen auftreten.

[0028] Furthermore, the symmetry of the circuit boards should also be considered. For example, simply or doubly symmetrical circuit boards, such as square and / or trapezoidal boards, can be used. Each circuit board can be assigned a vector that indicates the longitudinal direction of the strip or coil from which the respective circuit board is derived. The tip of the vector is oriented towards the beginning of the strip, describing the edge of the strip or coil that is visible from the outside before unwinding and thus runs parallel to the strip's transverse direction. Based on the vector, it is therefore possible to trace how the material forming a circuit board was originally oriented on the coil or the unwound strip.

[0029] When a stack of singly or doubly symmetrical blanks is fed into a press, the vector can point in either the direction of material flow or the opposite direction. The direction of the vector depends solely on how the stack is fed into the press or production line. This direction is generally random. If the properties of the semi-finished product or the material itself are not symmetrical with respect to the axis of symmetry, which runs parallel to the strip's longitudinal direction, discontinuity can also occur. This is particularly relevant for coils that are split in the middle, as such coils typically exhibit a non-symmetrical and non-constant distribution of the semi-finished product properties across the strip width, as can be the case, for example, with trapezoidal blanks.

[0030] Furthermore, discontinuities in the properties of a circuit board can also arise if the corresponding coil is joined together by welding several different strips of the semi-finished product. In this case, the properties of the circuit board located behind the weld seam can differ from those of the circuit boards located before the weld seam. The joining of the strips by a manufacturer of the semi-finished products is described in 23-3760 PIF.

[0031] 6 is usually tolerated by the operators of the press shops, although this represents a process change and may lead to downtime in the process.

[0032] Furthermore, discontinuity can occur, for example, when a robot arm is used to insert circuit boards into a press, alternately feeding boards from two stacks of the production system. If the circuit boards in the respective stacks have different properties, this case is particularly unfavorable, as discontinuity can occur during the transition from the first to the second stack, and thus with each change.

[0033] The inventive method now makes it possible to avoid component qualities that lead to defective components in the event of discontinuities occurring, by evaluating the defined sequence of the stacks or circuit boards to be processed and, if necessary, issuing a corresponding warning.

[0034] In other words, the process includes a discontinuity check at every transition from one circuit board to another, or from one stack to another. This means that discontinuities within a stack are also examined. This examination can be carried out, for example, once recorded board properties or the respective semi-finished product properties of the circuit boards in a stack are made available in a database. It should be noted that it is generally assumed that the properties of all processed semi-finished products meet the respective specifications.

[0035] This procedure can be carried out as soon as the processing sequence for the stacks is known. The sequence can be determined, for example, after the cut sheets have been stored. It is also possible that the sheets are delivered directly to the press or production line after cutting, and therefore the processing sequence is determined shortly before processing. In this case, the discontinuity check should be performed immediately before processing the sheets. In the worst-case scenario, no processing sequence is determined, or the sequence is changed at short notice. To account for this scenario as well, an additional discontinuity check can be performed directly during material processing, for example, as soon as a second stack is detected that follows the first in the processing sequence.This recognition can be achieved when feeding the respective stack, for example by means of an identification number of the stack, 23-3760 PIF.

[0036] 7 or its circuit boards into the press or production plant and / or at the latest when reading the serial number of, for example, the top circuit board of a stack. Thus, in the worst case, the sequence can only be determined after the processing of the circuit boards has already begun.

[0037] It can be advantageous, especially during the circuit board cutting process or after the completion of a stack of circuit boards, to calculate all possible combinations of this stack, for example with all previously cut stacks, and thus to have the evaluation of the discontinuity available for all combinations of discontinuity as soon as possible.

[0038] When transitioning from one circuit board to another or from one stack to another during processing in the manufacturing plant, several cases should be distinguished for the discontinuity assessment procedure.

[0039] To classify the identified discontinuity, a property value can be used. This value might describe a specific aspect of the discontinuity; for example, if the discontinuity is a semi-finished product property such as sheet thickness, the property value could describe the difference in sheet thickness between the sheet before and after the discontinuity. Historical data refers specifically to data stored in the database and used, for example, in previous production orders. A given data record of historical data might include, for instance, the component quality of a component, along with the associated semi-finished product properties of the underlying material, as well as the process parameters and / or state variables used in its production, such as the material temperature during processing.Historical data can be used, for example, to determine how different semi-finished product properties affect the respective component quality, even with identical process parameters. This allows for the classification of inconsistencies, as it can be determined whether the inconsistency is significant enough to cause the component quality to deviate from the required quality.

[0040] In the fourth step, the instruction signal can then be adjusted accordingly. For example, the instruction signal could indicate that a process parameter change should be performed due to the discontinuity. (See also 23-3760 PIF)

[0041] 8. The instruction signal can contain suggestions for corresponding process parameters that should be adjusted to achieve the required component quality during manufacturing. Furthermore, the instruction signal can, for example, only warn of a discontinuity that has not yet occurred and for which no historical data exists, indicating that a process parameter change might be advisable. In addition, the instruction signal can also indicate that a process parameter change is not necessary. Furthermore, a new or alternative sequence for processing the circuit boards and / or stacks can be specified, particularly to avoid or minimize the discontinuity. In some cases, it may be simpler to change the sequence than to implement a process parameter change.The instruction signal can also contain corresponding control signals for individual components of the production plant, so that the process parameters can advantageously be set independently or automatically.

[0042] The method according to the invention thus offers, among other advantages, the benefit of keeping the number of defective components of inferior quality particularly low.

[0043] In an advantageous embodiment of the invention, the instruction signal comprises a warning signal and / or a request to change process parameters and / or at least one process parameter that should be changed, wherein a corresponding value for the process parameter can be suggested or specified. Additionally or alternatively, the instruction signal comprises an updated processing sequence, wherein the processing sequence can be adjusted in such a way as to avoid and / or at least mitigate the identified discontinuity by processing the stacks and / or circuit boards in the new sequence suggested by the instruction signal. In other words, the instruction signal comprises all the information necessary for a plant operator to react appropriately to the discontinuity, for example, to perform the process parameter change.This offers the advantage that the instruction signal can be used particularly advantageously for operating the production plant.

[0044] In an advantageous embodiment of the invention, the expected component quality of the component, which follows the discontinuity on the circuit board, is estimated based on the historical data, and the instruction signal is adjusted depending on the expected 23-3760 PIF

[0045] 9

[0046] Component quality is output. In other words, historical data is evaluated if it includes component qualities corresponding to the respective semi-finished product properties and / or process parameters. This allows the necessary process parameter changes to be advantageously determined and incorporated into the instruction signal. This results in the advantage of a particularly precise execution of the process.

[0047] In a further advantageous embodiment of the invention, when manufacturing the component from the circuit board, the component quality is determined based on a vector characterizing the component, taking the instruction signal into account. In other words, properties such as shape, size, or the like are summarized as a vector in a vector space, which can, for example, be considered a quality space, and the component quality is determined, for example, by a difference vector from a target vector. This offers the advantage that the component quality can be determined particularly efficiently and used to generate new historical data.

[0048] In a further advantageous embodiment of the invention, historical data is stored in a database. During the production of the component from the circuit board by the manufacturing system, the database is continuously expanded, particularly through the recording of component quality. In other words, during operation of the manufacturing system, all variables, such as process parameters and associated component qualities, are continuously recorded, for example, using suitable sensors. This data is then used for historical data, for instance, for repeating the process. This allows, for example, the recording of newly occurring continuities and their effects on component quality, which can then be used for future instruction signals. This offers the advantage of enabling particularly precise execution of the process.

[0049] In a further advantageous embodiment of the invention, a vector is created for each circuit board based on its respective at least one semi-finished product property to determine the discontinuity, and vectors of successive circuit boards are compared. In other words, several semi-finished product properties (or at least one semi-finished product property) of the respective circuit boards are combined into a vector whose dimension corresponds in particular to the number of semi-finished product properties. Thus, it can happen that several values ​​are determined for one semi-finished product property and are included in the 23-3760 PIF.

[0050] 10

[0051] Vectors are stored. Vectors of two consecutive or generally sequential circuit boards can be compared, which advantageously allows discontinuities to be identified. This results in the advantage that discontinuities can be determined particularly precisely and / or easily.

[0052] In a further advantageous embodiment of the invention, at least one property value is derived from the comparison of the vectors, in particular using a metric. For example, an average distance between the vectors in the corresponding vector space is determined. This distance can represent the property value by which the discontinuity can be classified using vectors. This offers the advantage that the method can be carried out particularly efficiently.

[0053] In a further advantageous embodiment of the invention, the vectors of a number of circuit boards from a first stack are compared with the same number of subsequent circuit boards from the following stack to determine the property values. In other words, average values ​​are calculated over several circuit boards, particularly those exhibiting similar semi-finished product properties, and these are compared with a mean, median, and / or another statistical measure based on the same number of circuit boards, but potentially with different semi-finished product properties due to their stack affiliation. This offers the advantage that individual outliers with deviating semi-finished product properties can be identified, thus avoiding unnecessary process parameter adjustments.

[0054] In a further advantageous embodiment of the invention, the sequence of the circuit boards is determined by recording their serial numbers in the order they are taken for processing, and thus ultimately in the order in which they are intended for processing. This can be particularly helpful if, for example, no database entries were recorded during the stacking process, so that any discontinuity can be determined, especially with regard to the next circuit board. In other words, the sequence is determined when the circuit boards are fed into the production system and their serial numbers are recorded. This offers the advantage that the process can also be carried out in real time.

[0055] In a further advantageous embodiment of the invention, particularly during the formation of at least one stack of circuit boards, possible processing sequences are determined for the circuit boards intended for processing and / or the at least one stack of circuit boards, and discontinuities to be expected for each sequence are determined. 23-3760 PIF

[0056] 11

[0057] In other words, especially when the raw material—the circuit boards that will be processed in a subsequent order, for example—is already known, all possible permutations into which the circuit board sequence can be incorporated in the production line are tested, and any discontinuities are identified. This offers the advantage that the process can be used particularly quickly and / or efficiently.

[0058] In a further advantageous embodiment of the invention, the processing sequence with the fewest discontinuities and / or the lowest inconsistency is defined. In other words, a change in the processing sequence can be derived from the evaluation of the discontinuities. Thus, in addition to adjusting the process parameters, the processing sequence can also be influenced by the method. The processing sequence of the batches can therefore be specifically optimized, particularly with the aid of a suitable algorithm. Within the framework of such optimization, the sequence that results in the fewest discontinuities affecting component quality can be determined. This offers the advantage that the production plant can be operated with particular efficiency.

[0059] In a further advantageous embodiment of the invention, the at least one semi-finished product property, a sheet metal covering, a lubricant quantity, a roughness, a tribological property, and / or an elastoplastic property is described. In other words, the property can be any parameter that can be used to characterize the corresponding circuit board. This offers the advantage that the method can be applied with particular flexibility.

[0060] A second aspect of the invention comprises a computer program. The computer program can, for example, be loaded into a memory of an electronic computing device and includes program means to execute the steps of the method when the computer program is executed in the electronic computing device.

[0061] Advantages and advantageous embodiments of the first aspect of the invention are to be regarded as advantages and advantageous embodiments of the second aspect of the invention, and vice versa. 23-3760 PIF

[0062] 12

[0063] A third aspect of the invention relates to an electronically readable data carrier. The electronically readable data carrier comprises electronically readable control information stored thereon, which includes at least one computer program as just presented and is designed such that, when the data carrier is used in an electronic computing device, it can execute a method presented here according to the first aspect of the invention.

[0064] Advantages and advantageous embodiments of the third aspect of the invention are to be regarded as advantages and advantageous embodiments of both the second and the first aspect of the invention, and vice versa.

[0065] A fourth aspect of the invention relates to a manufacturing plant in which circuit boards can be processed into components as semi-finished products with certain process parameters and which can be operated by means of a method according to the first aspect of the invention.

[0066] Advantages and advantageous embodiments of the fourth aspect of the invention are to be regarded as advantages and advantageous embodiments of the first, second and third aspects of the invention, and vice versa.

[0067] For this reason, the advantages and beneficial further developments of the computer program and data carrier according to the invention are not described again here.

[0068] Further features of the invention will become apparent from the claims, the figures, and the description of the figures. The features and combinations of features mentioned above in the description, as well as the features and combinations of features mentioned below in the description of the figures and / or shown in the figures alone, are not only usable in the combinations specified, but also in other combinations or on their own.

[0069] The invention will now be explained in more detail with reference to a preferred embodiment and the drawings. The drawings show:

[0070] Fig. 1 shows a schematic flowchart for a method for operating a manufacturing plant; and

[0071] Fig. 2 shows a schematic view of the production facility. 23-3760 PIF

[0072] 13

[0073] Figure 1 shows a schematic flowchart for a method for operating a production plant 1, in particular for specifying a process parameter change for the production plant 1, in which circuit boards 2 or stacks 3 of circuit boards 2 are processed from semi-finished products with certain process parameters into components 4.

[0074] The inventive method comprises the following steps S1 to S4:

[0075] In step S1, a sequence is determined in which the circuit boards 2 and / or the stacks 3 are intended for processing in the production plant 1.

[0076] In step S2, a discontinuity between successive boards 2 and / or stacks 3 is determined based on at least one half-life property of the respective board 2 and / or a parameter and / or an environmental state.

[0077] In step S3, the identified discontinuity is classified based on at least one property value and / or historical data.

[0078] In step S4, an instruction signal is output, for example to a display device or to a control unit 5, which is designed to control the production plant 1, depending on the classification of the detected discontinuity.

[0079] Figure 2 shows a corresponding production plant 1 with two stacks 3, each containing several circuit boards 2. The production plant 1, which can be designed as a press, produces a component 4 from each circuit board 2, for example by forming. The production plant 1 shown can be operated, at least partially, by a control unit 5. The control unit 5 can, in particular, include an electronic computer.

[0080] Depending on the processing of the circuit boards 2, for example during cutting in a coil line and during the bundling of the circuit boards 2 into stacks 3, discontinuities can occur between two circuit boards 2 and especially between two stacks 3, which can be identified based on at least one semi-finished product property or the deviation of the corresponding semi-finished product property. 23-3760 PIF

[0081] 14

[0082] To determine the processing sequence, the serial numbers of the individual circuit boards 2 are recorded, for example. It is fundamentally conceivable to calculate all possible combinations of the stack 3 with all previously formed and / or stored stacks 3 immediately during the cutting process of the circuit boards 2 or after the completion of a stack 3, and to record any discontinuities that may occur due to the properties of the semi-finished product.

[0083] When discontinuities occur, several cases must be distinguished, each of which could have been recorded in the historical data or at least be derived from it:

[0084] Thus, a change in the semi-finished product properties following a discontinuity can lead to a violation of the quality criteria or component quality of the produced component(s), even if the process parameters remain unchanged. This event demonstrates that a discontinuity in the semi-finished product properties results in a violation of the quality criteria of the corresponding components. It can be assumed that a plant operator or a process control system (e.g., using the electronic computer) will adjust the process parameters after this event to ensure that the produced components meet the quality requirements. The adjustments to the process parameters following the discontinuity therefore provide guidance for the future on how to respond to such discontinuities with a change in process parameters.

[0085] The selection of process parameters can be determined using suitable algorithms and provided to the plant operator via an assistance system and / or a display device, for example, the control unit 5. Additionally or alternatively, the algorithms could directly initiate an automatic change of the process parameters by the production plant 1 or the control unit 5. The process parameters include, in particular, all parameters that can be changed on tooling, transport systems, presses, and / or peripheral devices, for example, on a re-lubrication system, in order to influence the quality of the manufactured components 4.

[0086] Furthermore, it could happen that, due to the inconsistency, the quality of the corresponding components 4 just barely meets the quality criteria. In such a case, 23-3760 PIF

[0087] 15

[0088] Given this situation, it is expected that the process parameters will not change. If this is classified or determined, a warning should be generated using the instruction signal.

[0089] In another case, the process parameters could be changed immediately upon the occurrence of a discontinuity, so that no violation of the quality criteria of the produced components 4 occurs. From this event, it can at least be deduced that a reaction or adjustment of process parameters was probably necessary to achieve the required quality of the produced components 4.

[0090] Furthermore, in the event of discontinuity, it is possible that the discontinuity will not affect the quality of component 4 at all, which should also be recognized. This could lead to the elimination of the need to define a processing sequence, which entails a certain amount of effort in production.

[0091] The four cases mentioned can be derived from historical data. The following section describes how discontinuity can be classified in particular:

[0092] Classification occurs before the instruction signal is issued. A comparison with historical data determines whether similar discontinuities have occurred in the past. For example, if it is determined that a discontinuity arising from the identified sequence will not result in a violation of the quality criteria for component 4, a warning about such a discontinuity via the instruction signal can be omitted. If the current discontinuity has not yet occurred, a warning is issued. However, it can be indicated that it is not yet known whether the discontinuity will lead to a violation of quality criteria.If historical data indicates that the inconsistency will lead to a violation of quality criteria unless a corresponding reaction or adjustment of process parameters is made, a warning is issued and / or, optionally, a recommendation is given to the operating personnel of the plant, or corresponding process parameters, such as those determined by algorithms, are stored in the instruction signal.

[0093] After processing the circuit boards 2, a reassessment of the discontinuity is carried out, checking to what extent the discontinuity actually leads to a violation of 23-3760 PIF.

[0094] 16

[0095] This has led to adjustments in quality criteria and / or process modifications. The results of such an analysis are written to the database along with further measurement data, allowing the historical data set to be continuously expanded, particularly with each production run of component 4. This allows the results of the re-evaluation to be used for future discontinuities, whereby the aforementioned case distinctions can be made during this evaluation.

[0096] Furthermore, the classification should be achieved in particular by quantifying the discontinuity of the semi-finished product properties. The properties of a circuit board 2 can, for example, be summarized as a vector. Each component of this vector, which is referred to below as a property vector, describes a single property or semi-finished product property, such as the amount of lubricant at a specific location on the circuit board 2. The vector can summarize several properties, such as sheet thickness, lubricant quantity, roughness, and / or elastoplastic material properties. To be able to evaluate corresponding differences in the properties of the circuit boards 2, the property vectors can, for example, be interpreted as position vectors.This results in two points in the property space when comparing two circuit boards 2, and the Euclidean distance between these points can be used as a measure of the difference in the properties of the two circuit boards 2. This difference could, in turn, serve as a measure of the discontinuity of the properties of circuit boards when switching between stacks 3, for example, by comparing the bottom circuit board 2 of the first stack 3 with the top circuit board 2 of the second stack 3. To make the evaluation of this difference in properties between two stacks 3 robust against measurement artifacts, the corresponding property values ​​of a point cloud in the property value space could be derived from all i bottom circuit boards 2 of the first stack 3. Similarly, a second point cloud could be generated that represents the properties of the j top circuit boards 2 on the second stack 3.Subsequently, the center of gravity for both point clouds could be calculated, and, for example, the Euclidean distance could be used to determine the distance between these two centers of gravity.

[0097] This distance can be interpreted as a measure of discontinuity. For example, i=j can be chosen, and in particular, 102 boards can be included. Furthermore, other metrics for determining the distance are conceivable, such as a Manhattan distance, where the summation of the 23-3760 PIF is performed.

[0098] 17 absolute differences of the individual coordinates of the points. Additionally, a weighting of the individual quantities under consideration can also be applied.

[0099] The method presented here can be used to evaluate discontinuities within a stack of 3. Any number of other metrics can be used to quantify this discontinuity. For example, if the properties or semi-finished product properties are Boolean values ​​and / or categorical quantities, the distance measure must be adjusted accordingly.

[0100] In addition to or as an alternative to the quantification of semi-finished product properties shown so far, an absolute level of semi-finished product properties could also be considered. For example, two discontinuities whose measure of discontinuity has the same or a similar value can nevertheless have different effects on the quality of the produced components 4.

[0101] For example, if the sheet thickness of the first stack 3 is at the midpoint of the specification, the discontinuity with the sheets 2 of the second stack 3 might have a less critical impact on the quality of the components 4 to be produced than if the sheet thickness of the first stack 3 were at the edge of the specification. In this context, it is possible that the discontinuity, to supplement the position vector of the first point cloud, describes the properties of the sheets 2 before the discontinuity occurs, in terms of the temporal sequence of the processing of the sheets 2. Thus, a discontinuity can be characterized by the difference in the properties of the semi-finished product, as well as the properties of the semi-finished products themselves.Similarly, in addition to the semi-finished product properties, the discontinuity analysis can also take into account process parameters and / or environmental conditions during processing, for example in the cutting system, during the cutting of the circuit boards 2.

[0102] The impact of a discontinuity on the quality of the produced components 4 can be easily assessed if process parameters n parts before and k parts after the occurrence of the discontinuity were constant in the past. This means that the change in the quality of the produced components 4 in the past can be directly attributed to the occurrence of a discontinuity in the semi-finished product properties resulting from the processed circuit boards 2. Based on this information, it can be deduced to what extent a discontinuity has led, or will lead, to a violation of the quality requirements of the produced components 4. (This was discussed in 23-3760 PIF.)

[0103] 18

[0104] In the past, if process parameters were adjusted or a process parameter change occurred when a discontinuity occurred, this can be interpreted, for example, as a clear indication that it is necessary to react to such a discontinuity in order to meet the quality requirements for the produced components 4.

[0105] If at least two stacks 3, as shown in Figure 2, are available, it may be necessary, for operational reasons, to take the circuit boards 2 alternately from one of the two stacks 3 when feeding them to the production system 1. It should be considered that in this case, if the discontinuity (between the two stacks 3) could lead to a violation of the quality criteria of the produced component 4, it will not be efficiently possible to individually set different process parameters or perform the process parameter change for each component 4 during this transition phase, even though this would be necessary to ensure that all circuit boards 2 processed in the transition phase result in components 4 that meet the quality requirements. To avoid this problem, it is particularly important to ensure that discontinuities between stacks 3 are avoided as much as possible.However, inconsistencies in stack 3 that occur outside the transition phase are not affected, as these can be addressed relatively easily.

[0106] To quantify quality differences of components 4, the invariance of the quality of the produced components 4 with respect to the occurrence of discontinuities can be investigated. For example, the quality of the i components 4 resulting from the i circuit boards 2 of the first stack 3 could also be represented as a vector. This vector could, for example, contain a deviation of the geometry of the respective component 4 from the target geometry. Other quality characteristics, such as cracking, wrinkling, and / or similar features, could also be included in such a vector. Similarly, the quality of the j components 4 from the j circuit boards 2 of the second stack 3 could also be stored. Again, it would thus be possible to calculate a difference vector between point clouds that describes a quality difference. Furthermore, it can be assessed whether a quality difference is significant.For example, if a discontinuity leads only to a minor difference in quality, this discontinuity should not be considered critical. To determine the significance of the quality difference, the quality difference can also be viewed as a vector in a quality space. This quality space can be a space bounded by all quality criteria. The boundaries of the space are thus derived from individual 23-3760 PIF.

[0107] 19

[0108] Tolerance limits of the quality criteria are defined. Thus, it would be possible, for example, to relate each component of the quality difference vector to the distance between a lower and an upper limit, or to calculate the corresponding quotient. These quotients can then be combined, for example, using a maximum norm, into a scalar value, which is referred to as quality significance. If the quality significance were to take the value 1, then at least one of the components of the quality difference would be equal to the distance between the tolerance limits. A specific limit value could then, for example, be a quality significance of 0.1. This would mean that all components of the quality difference vector are less than 0.1 with respect to the distance between the tolerance limits, and therefore the impact of the discontinuity on the quality of the produced components is considered to be low.The given limit value is an example; appropriate values ​​should be chosen depending on the production process and its requirements. Thus, the quality significance may also be chosen differently for each parameter that describes the quality of component 4.

[0109] The process can employ a machine learning method (machine learning model) to predict the impact of discontinuities in semi-finished product properties on the quality of the manufactured components. Historical data can be used as a training dataset. This would be particularly advantageous for evaluating discontinuities that have not yet occurred.

[0110] Without such a machine learning model, it is particularly difficult to search for similar discontinuities that occurred in the past and to infer the extent to which the discontinuity is likely to affect component quality. The machine learning model could therefore replace the search for the most similar discontinuities from the past.

[0111] The amount of historical data should be particularly high to effectively train a suitable model. In practice, it can be useful to pursue both approaches in parallel and / or depending on the available data. The method presented here can be advantageously used to monitor the consistency of the properties of semi-finished products, taking into account the quality of the manufactured components. -3760 PIF

[0112] 20

[0113] Reference symbol list

[0114] Circuit board manufacturing plant

[0115] Stack of components, control unit, first step, second step, third step, fourth step

Claims

1. 23-3760 PIF 21 Patent claims 1. Method for operating a manufacturing plant (1) in which circuit boards (2) are processed as semi-finished products with specific process parameters into components (4), comprising the steps: Determining the sequence in which the circuit boards (2) are to be processed in the manufacturing plant (1); (S1) Determining a discontinuity between successive circuit boards (2) based on at least one semi-finished product property of the respective circuit board (2) and / or a parameter and / or an environmental condition of a cutting system; (S2) Classifying the identified discontinuity based on at least one property value and / or historical data; (S3) and - Output of an instruction signal to the production plant (1) depending on the classification of the detected discontinuity. (S4) 2. Method according to claim 1, characterized in that the instruction signal comprises a warning signal and / or a request to change process parameters and / or at least one process parameter and / or an updated processing sequence.

3. Method according to claim 1 or 2, characterized in that, based on the historical data, an expected component quality of the component (4) which is derived from the circuit board (2) following the discontinuity is estimated and the The instruction signal is output depending on the expected component quality.

4. Method according to one of the preceding claims, characterized in that, in the manufacture of the component (4) from the circuit board (2) taking into account the 23-3760 PIF 22 The instruction signal determines the component quality based on a vector characterizing the component (4).

5. Method according to one of the preceding claims, characterized in that the historical data are stored in a database, wherein the database is continuously expanded when the component (4) is produced from the circuit board (2) using the manufacturing system (1).

6. Method according to one of the preceding claims, characterized in that, to determine the discontinuity for each circuit board (2), a vector is created based on its respective at least one semi-finished product property and vectors of successive circuit boards (2) are compared.

7. Method according to claim 6, characterized in that the property value is derived from the comparison of the vectors, in particular on the basis of a metric.

8. Method according to claim 6 or 7, characterized in that, to determine the property value, the vectors of a number of circuit boards of a first stack are compared with the same number of subsequent circuit boards of the following stack.

9. Method according to one of the preceding claims, characterized in that Determining the order of the circuit boards is done by recording the serial numbers of the circuit boards in the order in which they are taken for processing.

10. Method according to one of the preceding claims, characterized in that for the circuit boards and / or the at least one stack intended for processing 23-3760 PIF 23 possible processing sequences for circuit boards are determined, and for each determined sequence, expected discontinuities are identified.

11. Method according to claim 10, characterized in that the sequence with the fewest discontinuities and / or the lowest discontinuity is determined as the processing sequence.

12. Method according to one of the preceding claims, characterized in that the at least one semi-finished product property describes a sheet thickness, a lubricant quantity, a roughness, a tribological property and / or an elastoplastic property.

13. Computer program which can be directly loaded into a memory of an electronic computing device (9), comprising program means to execute the steps of the method according to any one of claims 1 to 12 when the computer program is executed.

14. Electronically readable data carrier with electronically readable control information stored thereon, comprising at least one computer program according to claim 10 and designed such that, when the data carrier is used in an electronic computing device (9), it performs a method according to one of claims 1 to 12.

15. Manufacturing plant (1) in which circuit boards (2) can be processed as semi-finished products with specific process parameters into components (4) and which can be operated by means of a method according to one of claims 1 to 12.

Images