System for monitoring events, especially stops, for identifying causes, and for generating recommendations for troubleshooting, as well as computer program product and use

Abstract

System for monitoring at least one type of stop in a production line for at least partially automatic, preferably fully automatic processing of packaging material, in particular cardboard boxes, for detecting the errors / causes causing the at least one type of stop, and for preferably generating error-specific at least one visually and / or audibly detectable recommendation for troubleshooting, in particular for at least one skilled worker operating / monitoring the production line, wherein the respective production line is an openly accessible production line of, for example, over 10m in length with several functionally distinct process steps, each with manual intervention capability for manual troubleshooting in the corresponding area / production line section of the respective process step, wherein the system has a computing unit with data storage.which communicates with a plurality of sensors monitoring the functioning of the production line and a plurality of optical detection units, at least one of which can be arranged / is arranged on the corresponding production line section with a field of view from above, preferably oriented at least approximately vertically downwards; characterized in that the system is configured to record data concerning at least the movement path along at least two production line sections of at least one specialist supervising the production line and, in response to a stop, to correlate at least one temporally predefinable sequence of the specialist's movement path with at least one specific type of stop, preferably for the purpose of generating at least one action recommendation and for outputting at least one action request, e.g. in file form.preferably in video file format, and / or on at least one visual and / or acoustic display unit of the system.

Description

TECHNICAL AREA

[0001] The present invention relates to a system for monitoring at least one type of event, in particular the stoppage of a production line for processing packaging material, for detecting errors or causes that trigger at least one type of stoppage / event, and for generating at least one recommendation for correcting the error, wherein the system has a computing unit with data storage which communicates with a plurality of sensors monitoring the operation of the production line and a plurality of optical detection units, at least one of which can be arranged on the corresponding production line section with a field of view from above overlooking the corresponding production line section.Furthermore, the present invention also relates to a computer program product comprising commands which, when the computer program product is executed on a computer, cause the computer to perform steps for controlling / regulating a method for monitoring events, in particular stops, for detecting errors or causes, and for generating recommendations for action, particularly when optimizing processes for the automated processing of cardboard boxes.Last but not least, the present invention also relates to the use of automated data processing means for specifying the manner of acquiring and evaluating (video) data concerning processing operations of a production line for processing packaging material, and for implementing a fault analysis and troubleshooting guidance functionality in coordination with a central control system of the production line in response to a production line stoppage, and for issuing at least one action prompt for troubleshooting, or a corresponding implementation of a computer program product. In particular, the invention relates to a system according to the preamble of the independent claim and its implementation for a corresponding method. BACKGROUND OF THE INVENTION

[0002] Packaging materials such as cardboard boxes are typically processed on relatively long production lines, usually involving several consecutive, functionally distinct process steps such as cutting, folding, stopping and / or aligning, gluing, or similar operations. The material feed (infeed) can also be designed as a separate functional section. The respective production line, for example, comprising a folding carton gluing machine / system, is usually operated and monitored by skilled personnel, for instance, to ensure correct material feeding or batch alignment on the conveyor belt (e.g., overlapping or non-overlapping). Such production lines are deliberately designed by engineers to be as openly accessible as possible (at least from the sides) and to be adaptable, especially since packaging materials such as cardboard boxes are often difficult to process.Cardboard boxes, especially folding cartons, can vary considerably in design (keywords and components include: size adjustment, web adjustment, side aligners, feed belts, adjusting tongues, servo drive components, length monitoring, tilt adjustment, speed adjustment, multi-stage folding, press belt adjustment, pressure zones, and outfeed height adjustment), so the production line is typically set up for each specific order. Once the process is running, the operator can usually intervene at several points along the production line and, for example, correct the alignment of individual batches of packaging material. In the event of a production process error caused by a material or machine malfunction, the production line stops automatically or must be stopped manually.The specialist must then locate the fault as quickly as possible, identify the error as quickly as possible, and rectify it as quickly as possible, especially to minimize production downtime. It goes without saying that acquiring the necessary expertise requires a comparatively long training period for the specialist involved, possibly spanning several years. Conversely, it is understandable that commissioning a production line can require not just a few, but numerous iterations, for example, repeating a sub-step 20 times before it can be predefined as the standard for subsequent production runs. It is also understood that, depending on their experience, the specialist will be under considerable time pressure and potentially also physical and / or psychological stress during the troubleshooting process.If the fault is not found immediately on a production line longer than 15m, this may be a cause of further sources of error (now on the part of the specialist) and also of risks, especially with regard to occupational safety.

[0003] Based on this, there is interest in improved procedural framework conditions for error detection in connection with stops and for at least partially manual error correction immediately in response to a respective stop, and last but not least, interest in improved work ergonomics, i.e., improved working conditions for the specialists involved.

[0004] According to the state of the art, various types of sensors and surveillance cameras are already being used, particularly in the direct context of process optimization during production. For example, publication DE 603 ​​17 073 T2 describes a multi-camera inspection system in the context of machine control, specifically for quality monitoring during the ongoing process.

[0005] Based on the current state of the art, there is a need for systems and procedures that facilitate easier and, ideally, more time-efficient manual intervention for troubleshooting in at least partially automated processes for handling packaging materials. Furthermore, there is a particular interest in greater system support for manual troubleshooting and resolution in response to stoppages, especially with regard to relieving the workload of the skilled workers involved. SUMMARY OF THE INVENTION

[0006] The task is to provide a system for facilitating the detection and understanding of errors that cause, in particular, a stoppage of at least partially automated packaging material processing. This system should simplify and enable the manual intervention of at least one skilled worker in the process or a corresponding production line in a particularly efficient manner. Furthermore, the task is to design such a system, or its implementation for a corresponding procedure, in such a way that the identification of the boundary conditions that cause the stoppage by the respective skilled worker is facilitated, especially to reduce the situational difficulties that the skilled worker faces at the moment of the stoppage and before the goal of efficient error resolution. Finally, the task is to design such a system, or its implementation for a corresponding procedure, in such a way that the respective skilled worker can easily identify the boundary conditions that cause the stoppage.its implementation is designed in such a way that the information recorded in the context of the stop is processed in such a way that the specialist involved is enabled to rectify the corresponding cause of the error quickly, efficiently and, last but not least, in a manner that complies with occupational safety regulations.

[0007] This problem is solved by a system according to claim 1, by a production line equipped therewith, and by its implementation for a corresponding method, as well as by uses according to the corresponding dependent use claim. Advantageous embodiments of the invention are explained in the respective dependent claims. The features of the exemplary embodiments described below can be combined with one another unless explicitly stated otherwise.

[0008] A system is provided for monitoring at least one type of event, in particular a stop (temporary failure) of a production line for at least partially automatic, preferably fully automatic processing of packaging material, especially cardboard boxes, for detecting the errors / causes causing the at least one type of stop (or the corresponding event), and for preferably generating at least one visually and / or audibly detectable recommendation for action to correct the error, in particular for at least one specialist operating / monitoring the production line, wherein the respective production line is an openly accessible production line of, for example, over 10m in length with several functionally distinct process steps, each with the possibility of manual intervention for manual error correction in the corresponding area / production line section of the respective process step (or(the system is set up for such a production line), wherein the system has a computing unit with data storage which communicates with a plurality of sensors monitoring the operation of the production line and a plurality of optical detection units (or can be brought into communication with each of them, in particular wirelessly, e.g., via near-field communication protocols or the Internet), of which at least one can be / is arranged on the corresponding production line section with a field of view from above, preferably in an orientation at least approximately perpendicular downwards;

[0009] According to the invention, it is proposed that the system is configured to record data relating to at least the movement path along at least two production line sections of at least one skilled worker supervising the production line and, in response to a stop or the corresponding event, to correlate at least one temporally predefinable sequence of the skilled worker's movement path with at least one specific type of stop / event, preferably for the purpose of generating at least one action recommendation and outputting at least one action request, e.g. in file form, preferably in video file form, and / or on at least one visual and / or acoustic display unit of the system.This facilitates the rapid identification, understanding, and localization of the root cause of any error, enabling particularly targeted manual intervention by the relevant specialist. This promotes a highly efficient method of troubleshooting, allowing the production process to be restarted promptly (minimal downtime and only slight delays). Information regarding the stop, error, or event can also be processed by the system and effectively and clearly communicated to the specialist, either in the field (e.g., locally at the point of failure or at least in its vicinity) or in the form of prepared documentation (e.g., training materials, video files for training specialists). The downward projection enables particularly precise tracking of the movement.the movement pattern of a specialist, on the other hand the specialist can be recorded anonymously, in particular without necessarily having to provide anonymization system-side.

[0010] It has been shown that targeted monitoring of the measures typically taken by skilled workers in response to an unscheduled production line stoppage (or a comparable event) significantly facilitates training and further education of skilled workers and, consequently, considerably improves the (technical) support of the respective production line. The present invention is based on the understanding that it is of great importance to evaluate current troubleshooting practices by analyzing the measures typically taken by skilled workers and to identify optimization opportunities, whether for saving time or for more targeted error detection.In particular, with reference to the physical movements of the respective specialist, especially in a vertical top view (also anonymized) and with temporal restriction to a predefinable, comparatively small time window before / after the respective unscheduled machine stoppage (or comparable event), the system described here can, on the one hand, assist in the respective troubleshooting measure in real time (i.e., without significant time delays during the troubleshooting process), and on the other hand, training materials can be generated for the subsequent training of specialists based on experience gained directly in the field, without this entailing any noticeable editorial or documentation effort for the specialists or a unit operating the respective production line.

[0011] Insofar as the present disclosure refers to at least one type of stop, this is to be understood synonymously as a reference to a specific production line and an unplanned temporary failure of this production line or the production process, and therefore also to specific errors or causes. In other words, the implemented system described here for generating troubleshooting recommendations / requests advantageously relates the respective recommendation to a specific type of production line, to a specific type of stop, and thus also to a specific subprocess of packaging material production. Therefore, based on the concept described here, it can also be reproducibly applied, for example, to other processes.It can be used as a standardized training document for specific production lines to train or further educate skilled workers, for example, when a new production line is being commissioned for a new order production with skilled workers who are not yet familiar with it, possibly also with order-specific reference to certain types of packaging material (batches). However, it should be understood that the stop, or at least one type of stop, can also be a different kind of event than a stop, e.g., a comparable event such as sensor- and / or optically detected damage to a component of the production line and / or a component of the material being processed.While the implementation of the present invention is primarily intended for the purpose of evaluating technical relationships in the context of stops, a different type of event can also trigger the evaluation process described here, for example, any signal, any trigger from a sensor component or a control system, and not least, manual system input, e.g., to request documentation, training material (or instructions), and / or a call to action (for example, at the point in time when the setup of a production line for a new batch of packaging material has been completed, and / or a corresponding commissioning process has taken place). Therefore, in the following, references to a stop at least implicitly also refer to such a comparable event, which can be implemented as an equivalent trigger specific to each production line.

[0012] Insofar as the present disclosure refers to a recommendation for action, this generally means a measure suggested by the system, be it an immediate troubleshooting measure to be carried out in the field, or a recommendation for action within the framework of a training document.

[0013] Insofar as the present disclosure refers to troubleshooting, this must be understood to mean at least partially manual troubleshooting, i.e., a troubleshooting process that requires the intervention of a skilled person. In other words, the production line is not a fully autonomous production line, and the processing processes described here are not fully autonomous self-adjusting processes, but rather are open within the meaning of the present disclosure, thus requiring manual or human intervention from time to time. This is due in particular to the nature and individual shaping of the packaging materials to be processed, especially cardboard boxes.

[0014] Insofar as the present disclosure refers to a call to action, this is to be understood synonymously as a call to action issued via at least one display unit on a respective production line for immediate manual troubleshooting during the (stopped) production process, i.e. a kind of guidance service for the specialist involved on site, particularly in the area of ​​the fault location on the production line.

[0015] Where the present disclosure refers to the movement path of a skilled worker, this means, on the one hand, a movement of the skilled worker's entire body relative to the ground, i.e., a relative movement of the person as such projected onto the ground (e.g., the floor of a machine hall), and on the other hand, also a relative movement of at least one body part, e.g., an arm, of the skilled worker, e.g., relative to a machine / plant component, e.g., relative to a feeder or a path adjustment component. Thus, the (monitoring) system described here can be implemented for both a movement sequence of persons and a movement sequence of body parts, e.g., an arm or finger movement. Based on the present disclosure, the person skilled in the art can specify how and to what level of detail, e.g.,The movement path of the entire body and / or individual body parts is to be analyzed and evaluated, e.g., with regard to individual body movements (e.g., arm movement), as a function of the current relative position of the specialist relative to the corresponding production line component, depending on the phase.

[0016] Insofar as the present disclosure refers to monitoring of at least one type of stop or event for the detection of the at least one type of stop or event-causing error / cause, this is to be understood synonymously as at least the combination of measures proposed according to the invention: detection, correlation, generation, and output, or their interaction described herein. The detection described herein primarily refers to data collection regarding movements or movement patterns of the respective specialist involved; optionally, the detection can also include data collection concerning the manufacturing / processing process as such, although such data collection can also be carried out by a control system of the respective production line itself with appropriate metrological or...This can be accomplished using camera-based equipment (in this respect, data transfer or data exchange / comparison with the monitoring system described here is also possible). Alternatively, the data acquisition described here can be carried out, at least partially, using production line-specific acquisition technology (in particular sensors, camera equipment), thus also capturing the movements of an involved specialist; alternatively, the correlation can be carried out, at least partially, based on data provided by a production line-specific system and / or a centrally accessible server; alternatively, the output can be carried out, at least partially, using production line-specific output technology (in particular operator terminals, graphical user interfaces, display units).

[0017] Whenever the present disclosure refers to a machine or plant, this typically means a component of a production line, e.g., a production line section with a predefined function such as cutting, folding, or gluing. Whenever the present disclosure refers to a machine stoppage, this synonymously means a stoppage of a production line, a production line section, or a machine component of the production line, particularly due to a fault. However, the present invention also relates to implementations that use a comparable event or a manual request as a trigger to initiate the monitoring and evaluation process described herein.

[0018] Where the term user or operator is used in this disclosure, it is to be understood synonymously as a suitably involved specialist who receives support in troubleshooting on the respective production line by means of the system described herein.

[0019] Personalized terms, unless explicitly formulated in the neuter gender, may refer to all genders within the context of this disclosure. Any foreign-language expressions or abbreviations used here are standard industry terms and are familiar to those skilled in the art. Any synonymous German terms used / available may be indicated here in parentheses for the sake of completeness, or vice versa.

[0020] The present invention is also based on the concept of facilitating the interaction between humans and the production line by analyzing human behavior in response to a stop or event on the production line and using it as a basis for a recommendation or request for action, whether in the field in a given current situation on the production line, or in the context of further training and / or training materials for skilled workers who may be newly trained or still undergoing further training, e.g. also in the context of recruitment measures (other / new skilled workers are assigned to a specific production line or a specific packaging material processing process).As mentioned above, production lines for packaging materials, especially cardboard boxes, are usually set up and run on a case-by-case basis, so the present invention can potentially be used with each new order for contract manufacturing to prepare the employees or specialists involved. The largely open manufacturing processes described here, with manual intervention options at several (at least laterally accessible) sections, therefore differ fundamentally from, for example, a purely manufacturing process in the paper industry, where manual troubleshooting is not usually easily feasible (simply because the moving workpieces or paper rolls are far too massive to be manually aligned).

[0021] It is important to understand that the (monitoring) system described here can be implemented largely independently of a (central) control system for the manufacturing process itself. While the optimization of machine and plant components in industrial applications, particularly for the purpose of increasing process efficiency and performance, can be facilitated by the (monitoring) system described here, it can also be implemented independently for a given production line. In other words, the primary goal of the (monitoring) system described here is to analyze the behavior and actions of skilled personnel in connection with a stop or a predefined event and to generate a recommendation or action request based on this analysis. This recommendation or request facilitates troubleshooting, ideally tailored to the specific type of stop or event recorded.The system can be used to document events, whether directly during the troubleshooting process (at least in near real-time) or in the form of prepared training materials for subsequent training of the involved specialist or other specialists. In this respect, the system described here may also communicate with sensors monitoring the functionality of the manufacturing process, but primarily to signal the stop time and, if applicable, the type of stop to the system described here. Alternatively, the system described here can communicate directly with a (central) control system or a corresponding control device of the production line (or the corresponding functionally defined production line section) to obtain information regarding the time and type of the respective stop.event; in this case, the system described here communicates indirectly via the control system with sensors that monitor the functioning of the manufacturing process.

[0022] It is understood that the system described here can already generate the measures described here based on the data acquisition using the acquisition units described herein. Of course, the implementation can be made even more application-specific and more targeted with regard to process coordination if sensor data is also included in the analysis in addition to the data collected by the acquisition units. A person skilled in the art can decide for a given implementation whether and to what extent this is expedient for optimizing the potentially production line-specific functionality of the system described here. For example, at least one first sensor of the production line correlates with a first acquisition unit of the system (especially with regard to the same location or...).the same process of the corresponding processing method), and at least a second sensor of the production line with a second detection unit of the system, etc. This also enables a particularly lean implementation with at the same time a maximum broad data basis when coupling the system with a control device or a control system of the production line.

[0023] Examples of implemented / implementable sensors include: sensors for monitoring side aligners, feed (especially two-stage feed), inclination adjustment of conveyor belts, and / or (servo) drive sensors, path adjustment sensors, length monitoring sensors, speed sensors, pressure and temperature sensors.

[0024] The present disclosure reveals that the invention is situated within the context of processes requiring human monitoring and / or intervention. Skilled workers, as human instruments, cannot be replaced by robotics in the foreseeable future, at least not in the context considered here, where variable, flexible order-based production, even of relatively small batches, will remain of great importance. In other words, humans will continue to be a decisive factor with regard to efficiency and quality in the production lines considered here. Therefore, the manual actions performed by the skilled workers involved during production are of great importance, and any uncertainties or errors in this regard must be avoided as effectively as possible.The present invention also facilitates the implementation of the concept of recognizing, understanding, applying, thus directly supporting specialists in the problem / error correction process.

[0025] Exemplary key parameters for a typical manufacturing process of cardboard packaging are: processing width in the range of 1500 to 2500mm, longitudinal seam gluing at three, four or six points, machine speed in the range of 200 to 300m / min, press pressure setting adjustable in three zones, transfer stations between individual functional sections of the production line.

[0026] The present invention provides measures and aids for realizing, in particular, the following advantages, especially thanks to an advantageous combination of AI technology and real-time machine monitoring measures in the context of fault-related stops or similar events, for improving industrial packaging material processing processes, particularly also based on camera technology specifically implemented for individual processes of the production line with AI-supported (image) data evaluation: - Reduction of machine downtime, especially thanks to precise analysis and solution guidance; - automatic creation of training and optimization materials, especially in the context of error-related stops, particularly taking into account the behavior of specialists; - efficient use of AI tools or machine learning in the evaluation of image / video material and in the generation of recommendations for action, especially to support specialists or operating personnel and to maximize machine performance; - Ergonomic user guidance for particularly time-efficient, quick access to relevant information and instructions on the one hand and / or to the sources of error themselves, on the production line, on the other hand;

[0027] The present invention thus facilitates a particularly targeted awareness of errors, error sources, or error causes in direct temporal connection with (machine) stoppages or comparable events, and also enables the categorization of errors (or causes), thereby simplifying and making troubleshooting more efficient / effective. Furthermore, a specialist can be guided more or less in real time, based on system-generated recommendations for action.

[0028] For example, the implementation of AI models can also be carried out in such a way that the system records whether a specialist is moving in the right direction (especially on the way to the source of the error) and / or is performing the correct movements (especially in the area of ​​the source of the error for the purpose of troubleshooting, particularly by handling workpieces or packaging material), especially for the purpose of targeted troubleshooting.

[0029] Advantageously, system-generated solution proposals are visualized at least at one operator terminal on the respective production line. Preferably, at least two or three visualization points are implemented along each production line, each advantageously equipped with at least one display unit. Based on the present disclosure, a person skilled in the art can also equip each visualization point or display unit with an input function or user interaction interface, whereby the associated advantages may depend heavily on the type and extent of the respective production line. The visualization can optionally be carried out by means of two- or three-dimensional illustration or reproduction (output process).

[0030] The majority of optical detection units can ensure the collection of a data foundation upon which an advantageous AI-supported evaluation of (relative) movements can be carried out. This evaluation is similar to a situational analysis based on the relative movements of skilled workers or operators, with the aim of creating a data basis for the reproduction of image / video data and the resulting recommendations for action. Based on this, primarily visual assistance can be provided for troubleshooting (simultaneously with troubleshooting and / or as training material). The inclusion of movement data relating to the packaging material itself and / or to machine components (e.g., actuators, side aligners, or the like) is also possible, although this is not necessarily required for the implementation of the (monitoring) system described here.The latter type of motion data can also be captured or provided by the production line itself or by its control system.

[0031] For relatively complex production lines, at least seven or eight inspection points regarding potential sources of error may prove useful for implementing the (monitoring) system described here; however, there may also be significantly more or considerably fewer inspection points. A inspection point is not necessarily designed as a visualization point, or vice versa.

[0032] Thus, the present invention provides a system and its implementation for a corresponding method for monitoring and optimizing, in particular, machines in industrial applications for processing packaging material, wherein camera technology can be advantageously combined with AI-supported analysis, especially to improve the efficiency and performance of human-machine interactions. In particular, the following can be implemented: - Identifying and analyzing machine stoppages, determining the causes of errors for each interruption, especially also using AI models (AI-supported stoppage analysis and performance evaluation); - Live monitoring and video analysis, especially at critical positions of the respective machine / production line, particularly for providing live streams; - Viewing real-time videos, especially via a mobile application, an operator terminal and / or an admin platform; - Creating training and process optimization videos; - Possibility of signaling, via user input or button press, that video data of a machine setup process (or loading, running-in) should be recorded; - System-based automatic creation of training videos from advantageous camera perspectives, whereby the training videos can be used, for example, to train specialists for future orders; - Automatic creation of a video data-based guideline for similar future orders or processes, especially after reaching a desired machine performance (output threshold); - Issuance of recommendations for action (learning material and / or specific troubleshooting instructions) at advantageous positions on the production line, especially directly at / on a terminal; - Proximity- or position-dependent output of relevant videos or video data and instructions or recommendations for action and / or calls to action in the context of a respective stop that has occurred, especially as soon as a specialist or operator approaches a terminal (the terminal can be equipped with at least one proximity sensor), which further reduces the reaction time in troubleshooting and ensures particularly efficient problem solving or reactivation of a production process;

[0033] In the following, references to stops are also synonymous with comparable events, which may each be implemented as triggers specific to the production line, e.g., exceeding a threshold of a positional tolerance and / or an image deviation (e.g., AI-supported evaluation of image information regarding the target / actual condition of a cardboard box).

[0034] According to one embodiment, the system is configured, in response to a stop, to store the last recorded data concerning at least the motion path or the time-predefinable sequence of the motion path before / after the stop, correlate this data with at least one type of stop, and generate a prompt for action in the form of at least one corrective action. This prompt is then displayed on at least one visual and / or acoustic display unit of the system, particularly within five to ten seconds after the stop occurs. Advantageously, the visual and / or acoustic output is spatially predetermined within the area of ​​at least one affected production line section. This also facilitates particularly effective system support for highly efficient, at least partially manual, troubleshooting.

[0035] Advantageously, the system, or at least the production line itself, includes at least two or three display units distributed along the production line. This also facilitates easier identification of the location and type of defect by the respective specialist.

[0036] According to one embodiment, the temporally predefinable sequence of the motion path is defined within a time window of at least three seconds before the stop and / or at least five seconds after the stop. This allows, on the one hand, a simpler understanding of the boundary conditions leading to the error, and on the other hand, the analysis of the situation and the method of troubleshooting or the implementation of a corrective action can be focused on the particularly important time period immediately after the error occurs.

[0037] Of course, the time window can be extended, for example if the recommendation for action is also to be in the context of specific corrective measures or is to explain the latter, for example by referring to comparable errors that have already been corrected.

[0038] According to one embodiment, the system is configured to implement output based on a time-lapse, in particular by outputting the data last acquired in response to a stop, concerning at least the motion path or the corresponding predefined sequence of the motion path, as a video at a speed reduced by at least a factor of two, especially on at least one of the display units. This facilitates, among other things, the identification of error causes by specialists or operators, as well as an understanding of possible or actual sources of error. Such a time-lapse can also be described here as a slow-motion setting, for example, by playing back a time window of three to ten seconds for up to one minute.

[0039] According to one embodiment, the system is configured to generate at least one recommended course of action, at least partially, based on data stored in the data repository, particularly after correlation, for example, by referencing video data of previously successful troubleshooting measures on the same or a similar production line, especially for a comparable batch of packaging material. Such information retrieval from data repositories can also facilitate particularly rapid troubleshooting, especially by searching the system for stored recommended courses of action or video data on comparable error situations, particularly immediately in response to a stoppage.

[0040] The different types of stops can also be stored in a database of the system described here and / or a storage component of the respective production line itself and / or on a centrally accessible server or in a specially designated area of ​​the data storage of the system described here, particularly in the form of an error library, based on which the potentially possible causes of errors can be more easily identified and the recommendations for action to be generated can be created more effectively. The individual causes of errors can be characterized, for example, by a number of error characteristics, especially with reference to a specific production line and / or a specific batch of packaging material (e.g., a specific type of folding carton).

[0041] According to one embodiment, wherein at least one specific type of stop correlates with at least one defect characteristic, in particular at least one defect characteristic from the following group: defect position, defect time, batch identification characteristic, production line identification characteristic, functional processing step (e.g. feeding, gluing, cutting, printing, folding, transferring, aligning, or the like), spatial orientation of the workpiece, optionally packing or overlapping state of the workpiece with other workpieces; in particular specifically with reference to a particular production line and / or a particular batch of packaging material.This also facilitates a particularly systematic approach, especially in the sense of an iterative continuous improvement process, in which, if a similar error occurs again, the previously recommended or implemented measures are evaluated and taken into account in the new action recommendation. The correlation of stops with error characteristics can be implemented by the system described here, but can optionally also be implemented additionally or alternatively by the production line system. Alternatively, the error correlation can also be provided by the production line system itself, so that the system described here can generate the action recommendation / request based on this information.

[0042] It is understood that the system described here can be equipped with self-learning capabilities (in particular with at least one machine learning algorithm) and can be trained, especially with regard to previously recommended and / or actually implemented measures. In this respect, the system described here, or its implementation for a corresponding procedure, can also include a subsequent step following the generated action recommendation / request, in which actual troubleshooting measures taken in the field by the respective specialist (especially manual troubleshooting measures by specialists) are recorded and evaluated, particularly with the inclusion of artificial intelligence. In this way, action recommendations / requests to be generated can also be iteratively adjusted, in the sense of self-learning optimization, e.g.in the database described here, particularly depending on successful user-side actions. In this respect, the system described here can also be described as a dynamic system for generating recommendations / requests for troubleshooting. The recording and evaluation of actual troubleshooting measures taken in the field can also be implemented in the context of further human-machine interaction during the respective troubleshooting measure, especially with AI support.

[0043] According to one embodiment, the system is configured to acquire data after reaching a predefined target machine output or target production line output, particularly in a continuous manner, by temporarily buffering only a predefined historical time window of the acquired data and saving it permanently in the data memory only when the system stops (otherwise it is overwritten). By referencing the current output (e.g., a threshold value for an output parameter), a correlation can also be established in the context of processing speed, allowing for an evaluation of whether certain errors occur only within a specific range of output or only above a certain, relatively high output level.This also facilitates setting the highest possible output without excessively increasing the risk of errors (which may need to be individually recalibrated for each batch on a per-order basis). Furthermore, data can be recorded or initiated when a predefined target output range is exceeded, particularly for analyzing the causes of machine performance fluctuations.

[0044] According to one embodiment, the system is configured to perform at least the detection and, depending on the stop, also the correlation based on at least one performance / output parameter relating to the machining process of the respective production line, in particular depending on a fluctuation of the at least one performance / output parameter, especially when a predefined deviation from a target performance / output parameter is exceeded. This also facilitates proactive measures for the preventive handling of potential sources of error.

[0045] According to one embodiment, the system is configured to perform at least the correlation and / or at least the detection using at least one machine learning algorithm, or based on at least one machine learning algorithm, in particular based on at least one AI model implemented for the error detection described here. This also makes it possible to leverage further potential, especially in the context of video data analysis. For example, at least one machine learning algorithm is / will be implemented in the context of the step of determining the error type (especially for the purpose of quickly / targeted generation of a request for action) and / or in the context of the output step (especially for the purpose of effectively guiding the specialist along the production line or to the error location).

[0046] It is understood that the implementation of ML algorithms or AI models within the scope of the present invention may include a computer infrastructure or data processing architecture, particularly in the core of at least one of the computing units described here, which facilitates, enhances, or accelerates the execution of ML algorithms directly in / on the hardware (up to real-time processing) and / or makes it more energy-efficient, or at least partially enables it in the first place.For example, the computer-based and chip-based tools described here include at least one of the following components: photonic AI chips, particularly those with silicon photonic structures (a combination of electronic and optical data processing), optical waveguides at least partially replacing or supplementing electronic semiconductors, as well as multiplexing components, photon modulators, photodetectors, ring resonators, at least one dense wavelength division multiplexing (DWDM) component for the simultaneous processing of multiple data channels, at least one optical circuit integrated into at least one neural network (NN or DNN, deep neural network), and at least one photonic processor. For example, at least one NN and / or DNN is executed directly at the hardware level. As a result, particularly large datasets can be analyzed very quickly.For example, at least one photonic component is present in form printed directly onto a wafer, in particular at least one of the following photonic components: optical amplifiers, photonic circuits (PICs), polarization converters, splitters, optical waveguides, splitters, phase modulators.

[0047] According to one embodiment, at least one optical detection unit is arranged at each publicly accessible section of the production line with manual intervention for troubleshooting, preferably oriented at least approximately vertically downwards, for example at a height of three or four meters. This also enables location-specific, individually optimized data acquisition in the most relevant sections of the production line, particularly with regard to anonymization of involved users or specialists.

[0048] The aforementioned task is also solved by a computer program product comprising commands which, when the computer program product is executed on a computer (or a corresponding computing unit), cause it to perform steps for controlling / regulating the following method, namely a method for monitoring at least one type of event, in particular a stop (temporary failure) of an openly accessible production line for at least partially automatic, preferably fully automatic processing of packaging material, in particular cardboard boxes, for detecting the errors / causes causing the at least one type of event, in particular a stop, and for preferably generating error-specific at least one visually and / or audibly perceptible recommendation for action to correct the error, in particular for at least one specialist operating / monitoring the production line.wherein a plurality of sensors monitoring the operation of the production line and a plurality of optical detection units communicate with a computing unit with data storage, wherein the optical detection units have a top-down field of view overlooking a corresponding production line section; wherein data concerning at least the movement path along at least two production line sections of at least one skilled worker supervising the production line are recorded, and in response to an event, in particular a stop, at least a temporally predefinable sequence of the skilled worker's movement path is correlated with at least one specific type of event / stop, preferably for the error-specific generation of at least one action recommendation and for the output of at least one action request in file form and / or on at least one visual and / or acoustic display unit,for example, a display unit of the system described here; especially also when / through the use of a system described above. This results in the aforementioned advantages, particularly with regard to the highest possible degree of automated system support during the troubleshooting process, which is at least partially manual.

[0049] For example, the recommended action includes at least one fault message, e.g., in the following form: "The storage system is malfunctioning - please check the storage area." For example, the recommended action includes at least one error message, e.g., in the following form: "Machine stopped due to excessive workpiece width - please geometrically check and / or realign the workpieces." Such error information can also be location / time-specific or exhibit at least one other of the error characteristics described here.

[0050] According to one embodiment, in response to a stop, the last recorded data concerning at least the motion path or the time-predefinable sequence of the motion path before / after the stop are stored and correlated with at least one type of stop. A prompt for action in the form of at least one corrective action is generated and displayed on at least one visual and / or acoustic display unit or on the corresponding production line. The corrective action can be explained, for example, in the form of a video, advantageously also locally in an area of ​​the production line where the error occurred (optionally also centrally at a terminal). Simultaneously, a graphical user interface can be called up or automatically presented, particularly by the system, for example, when user input is useful for troubleshooting.concerning a relocation or repositioning of a machine component and / or the batch itself.

[0051] According to one embodiment, the temporally predefinable sequence of the motion path is specified within a time window of at least three seconds before the stop and / or at least five seconds after the stop, particularly at the system level. The time window can be comparatively small, which can facilitate focusing the troubleshooting and improve understanding of the error's origin, e.g., based on slow-motion playback.

[0052] According to one embodiment, at least one playback time window is specified for issuing the at least one action prompt, particularly by the system, and specifically at least one first playback time window for issuing a first action prompt and a second playback time window for issuing a second action prompt. This staggered timing can also promote understanding of at least one troubleshooting measure, for example, if the troubleshooting measure is to be carried out in several steps. The staggered timing thus also facilitates the guidance of a field technician.

[0053] According to one embodiment, at least one playback position for issuing the at least one action prompt is specified, particularly by the system, and in particular at least one first playback position for issuing the first action prompt and a second playback position for issuing the second action prompt. This also facilitates the implementation of a troubleshooting measure, especially when a fault is not only to be rectified very locally, but at several points along a comparatively long production line. The specialist involved can therefore also be guided visually / audibly to the specific location, e.g.by issuing a call to action in the corresponding section of the production line for a first corrective action further upstream of the production line (especially relative to the current position of the specialist involved), and by issuing a call to action in the corresponding section of the production line for a second corrective action further downstream of the production line.

[0054] The respective playback position is / will be advantageously defined relatively in length and width and optionally also in height relative to the respective production line.

[0055] According to one embodiment, the output is based on a time-lapse, in particular by outputting the data last acquired in response to a stop, concerning at least the motion path or the corresponding time-predefined sequence of the motion path, as a video at a playback speed reduced by at least a factor of two (here also referred to as slow motion). This facilitates a quick understanding of the cause of the error by the specialist involved, whereby the most relevant sequence can be selected and played back in time-lapse, particularly by the system itself.

[0056] According to one embodiment, the generation of at least one recommended action is based, at least in part, on data stored in the data memory, particularly after correlation. This also enables the preferably system-level automatic inclusion of empirical data and previously successfully implemented troubleshooting measures when generating action prompts, which may be particularly useful in the context of machine learning, especially for the iterative improvement of troubleshooting measures, particularly during the commissioning of new production lines or similar orders or comparable batches of packaging material.

[0057] According to one embodiment, data is acquired after reaching a predefined target machine output or target production line output, particularly in a continuous manner, by temporarily buffering only a predefined past time window of the acquired data and saving it only in the event of a stop or permanently storing it in the data memory. This also makes it possible to continuously ramp up the machining process to an output that still allows for a good compromise between efficiency and error risk.

[0058] According to one embodiment, generating at least one recommended course of action includes generating a fault description. This facilitates the classification of the fault situation by the respective specialist and also improves the comprehensibility of recommendations / requests for corrective action. In other words, the fault-specific correlation of machine stoppage and fault, preferably implemented system-wise, does not necessarily yet include a fault description that is readable / comprehensible by humans, especially since the latter can also depend on the nature of the cause.

[0059] According to one embodiment, at least the acquisition and, depending on the situation (stop-dependent), optionally also the correlation, is based on at least one performance / output parameter, in particular depending on a fluctuation of the at least one performance / output parameter, especially when a predefined deviation from a target performance / output parameter is exceeded. This also enables targeted analysis, particularly in phases in which the production line or process deviates from a target output. The at least one performance / output parameter can, for example, be linked to at least one motor or drive or its current power consumption.

[0060] According to one embodiment, at least the correlation is performed using at least one machine learning algorithm, or based on at least one AI model implemented for the stop-dependent error detection described here, particularly by training the corresponding AI model for the respective process to be monitored, either in the field or through production line-specific implementations. This also promotes a system-side training effect, especially in the case of recurring error causes, particularly during the commissioning of newly constructed production lines or newly implemented batches of packaging material.AI-supported generation of action recommendations, particularly in the context of AI-supported selection and presentation / output of video data relevant for specific error detection and resolution, can enable significant efficiency gains and, not least, facilitate a particularly targeted and ergonomic working method for specialists during extended periods of monitoring high-speed production lines. For example, at least one machine learning algorithm is / will be implemented in the context of determining the error type and / or in the context of generating the recommendations (optionally including system-based selection of the relevant sections of a video sequence) and / or in the context of outputting them.

[0061] According to one embodiment, the data acquisition is carried out using at least one machine learning algorithm, in particular based on at least one AI model implemented for recognizing the movement of a skilled worker, especially for evaluating the current movement of a skilled worker in the context of a fault-specific recommendation / request for action. This also enables particularly effective system support for the skilled worker involved in manual troubleshooting and fault correction, at least in near real time. In other words, the majority of optical detection units can be operated for video data acquisition in such a way that it can be evaluated whether a skilled worker is acting correctly in a given fault-solving situation, and if not, corresponding warnings or suggestions for improvement can be generated by the system, in particular in the sense of local guidance (e.g.,through local acoustic / visual indication) and / or expert guidance of the specialist, for the purpose of the most efficient and safest possible manual troubleshooting. At least one machine learning algorithm can facilitate or even enable the situation-specific evaluation of actions by an involved specialist with the desired or required efficiency and accuracy.

[0062] It is understood that the generation and / or output can also be performed using or based on at least one machine learning algorithm, particularly depending on specific user behavior, such as the current movements or movement patterns of an involved professional. In this way, the system can provide even more situation-specific, error-specific recommendations, especially in near real-time, e.g., with a delay of only milliseconds or a maximum of single-digit seconds, depending on the technically feasible response speed. In the context of each step of the monitoring described here (especially acquisition, correlation, generation, and output), a specific AI model can also be implemented, depending on the application-specific requirements.

[0063] The implementation of at least one machine learning algorithm described here can thus ensure at least one of the following technical effects: system-side identification / selection of the relevant time segment and / or image section of a video sequence; recognition of a specialist's movement patterns and immediate system-side output of instructions to prevent incorrect operation or unsuccessful troubleshooting attempts by the specialist involved; system-side comparison with a database with regard to the classification of a stop or...Errors as known (with corresponding predefined instructions for the specialist) or as novel errors and / or as novel attempts at error correction by the specialist involved (resulting in the need to generate a new recommendation for action for this specific error situation); (time-)efficient use of available resources in data collection and analysis in a comparatively complex context, particularly depending on current user behavior, especially simultaneously with the analysis of multiple parameters or (video) data.

[0064] The aforementioned task is also solved by a computer program product comprising commands which, when the computer program product is executed on a computer, cause the computer to perform steps for controlling / regulating a process according to the present disclosure on the computer, in particular a computer program product configured to specify the manner of outputting at least one recommendation for action when monitoring at least one type of event, in particular the stop of an openly accessible production line for at least partially automatic processing of packaging material, in particular configured to specify the manner of recording physical movements of at least one skilled worker and to specify times and / or positions when generating and outputting at least one recommendation for action for troubleshooting, in particular in response to and in temporal relation to an event, in particular a stop.Based on the aforementioned advantages, this also enables simple implementation specific to production lines, processes, or batches, for example, user-specific customization of the input / output method, e.g., via a graphical user interface, for example, provided via an application on an end device.

[0065] The aforementioned task is also solved by using a system in communication with a plurality of sensors monitoring the operation of an openly accessible packaging material production line and with a plurality of optical detection units, comprising a computing unit, data storage, and a communication module for monitoring at least one type of event, in particular a stop of the production line, for detecting the errors / causes causing the at least one type of event / stop, and for generating at least one visually and / or audibly perceptible recommendation for action to correct the error for at least one specialist supervising the production line, wherein the optical detection units are / will be arranged along the production line so as to provide a view of a corresponding production line section, and wherein data concerning at least the movement path along at least two production line sections are recorded by at least one specialist.wherein, in response to a stop, at least a temporally predefinable sequence of the movement path of the relevant specialist, in particular a progression over several seconds immediately after the stop, is correlated with at least one specific type of event / stop, in order to generate at least one action recommendation and to issue at least one action request; in particular in an implementation for a method according to the present disclosure, and especially when using a system according to the present disclosure. This allows the aforementioned advantages to be realized, particularly also with regard to the additional equipping of existing production lines with the system described herein.

[0066] The aforementioned problem is also solved by using a computer program product to specify the manner of acquiring and evaluating (video) data concerning processing processes of a production line for processing packaging material, in particular a computer program product according to the present disclosure, and to implement a fault analysis and troubleshooting guidance functionality in coordination with a central control system of or for the production line in response to a stop of the production line, and to output at least one action prompt for troubleshooting, wherein the output further includes at least one output of a video concerning a time window from a few seconds before the stop to a few seconds after the stop.The event encompasses, in particular, a video containing comprehensive image data relating to the development of the fault cause or the temporal progression of machine processing at the fault location, as well as comprehensive image data relating to the movement path or a predefined sequence of the movement path of the relevant specialist, especially immediately after the event, particularly after a stop. This allows the aforementioned advantages to be realized, especially with regard to retrofitting the functions or system described here into already designed, installed, and / or commissioned production lines, as a retrofit option in communication with or for the respective central control system of the production line. "A few seconds" preferably refers to a time window in the single-digit second range, e.g.,three or five seconds before / after the event to be evaluated on the production line (especially stop time or cause of error).

[0067] Summary: Automated processes for handling packaging materials are prone to errors and subsequent production line stoppages, resulting in temporary disruptions to workflows and output. For example, cardboard boxes requiring custom processing in relatively small batches have numerous parameters and settings that can introduce potential sources of error, especially when the process needs to be run for a short time or when batch-specific expertise is limited. Therefore, the personnel involved must strive to resolve any processing stoppages as quickly as possible.which is particularly not trivial in the case of long production lines and multi-stage, functionally distinct process steps. According to the invention, a system is provided for monitoring at least one type of event, in particular a stop of an open production line for at least partially automated processing of packaging material, for detecting the errors or causes triggering the at least one type of event / stop, and for generating at least one recommendation for troubleshooting for at least one skilled worker operating the production line. The system comprises a computing unit with data storage that communicates with a plurality of sensors monitoring the operation of the production line and a plurality of optical detection units, at least one of which is arranged on the corresponding production line section with a top-down view. The system is configured toTo record data concerning at least the movement path along at least two production line sections of at least one skilled worker supervising the production line and, in response to an event, in particular a stop, to correlate at least one temporally predefinable sequence of the skilled worker's movement path with at least one specific type of event / stop, preferably for the purpose of generating at least one action recommendation specific to the fault and for outputting at least one action request, e.g., in file form, preferably in video file form, and / or on at least one visual and / or acoustic display unit of the system. This allows skilled workers to be instructed in the field almost in real time, for particularly efficient fault correction.and / or training material or training for specialists can be provided before, after, or in connection with the commissioning of a production line. The invention also relates to an implementation of components of such a system for a corresponding method for monitoring an event / stop, for identifying (error) causes, and for generating at least one recommendation or request for action. BRIEF DESCRIPTION OF THE FIGURES

[0068] The invention is described in more detail in the following drawings, whereby reference numerals not explicitly described in a particular drawing are made to the other drawings. They show: Fig. 1A, Fig. 1B in a top view and in a side view a system implemented on a production line according to an exemplary embodiment; Fig. 2A, Fig. 2B, Fig. 2C, Fig. 2D, Fig. 2E, Fig. 2F in a perspective side view and in several perspective detail views a system implemented on a production line according to an exemplary embodiment; Fig. 3A, Fig. 3B, Fig. 3C, Fig. 3D, Fig. 3E, Fig. 3F again in a perspective side view and in several perspective detail views, but from a different perspective than in Fig. 2, a system implemented on a production line according to an exemplary embodiment; Fig. 4A, Fig. 4B, Fig. 4C, Fig. 4D, Fig. 4E, Fig. 4F according to the perspectives from Fig. 3 the field of view of one / the respective detection unit of a system according to exemplary embodiments, wherein the production line itself is only very slightly indicated in a schematic way, for the purpose of better visibility or Recognizability of the fields of view of the individual detection units; Fig. 5A, Fig. 5B each in perspective side view in exploded view a holder for at least one detection unit of a system according to embodiments; Fig. 6A, Fig. 6B which are in the Fig. 1A, Fig. Components shown in 1B, supplemented by motion path representations; Fig. 7A, Fig. 7B, Fig. 7C, Fig. 7D, Fig. 7E, Fig. 7F, Fig. 7G which are in the Fig. Components shown in 2A to 2F, supplemented by motion path representations; Fig. 8 in schematic representation steps of a procedure for which one / the can be implemented according to exemplary embodiments; Fig. 9 in a side view in schematic representation a system according to exemplary embodiments, or a production line equipped with it; DETAILED DESCRIPTION OF THE FIGURES

[0069] The invention will first be explained with general reference to all reference numerals and figures. Specific features or individual aspects, or aspects of the present invention that are clearly visible / representable in the respective figure, will be addressed individually in connection with that figure.

[0070] A system 100 is provided for monitoring at least one type of stop of a production line 1 for at least partially automatic, preferably fully automatic processing of packaging material 3, in particular cardboard boxes, for detecting the errors / causes causing the at least one type of stop, and for preferably generating error-specific at least one visually and / or acoustically detectable action recommendation 5 for troubleshooting, in particular for at least one skilled worker 7 operating / monitoring the production line, wherein the respective production line 1 is an openly accessible production line of, for example, over 10m in length with several functionally distinct process steps, each with manual intervention options for manual troubleshooting in the corresponding area / production line section 1.1 of the respective process step, wherein the system 100 has a computing unit 10 with data storage, which communicates with a plurality of sensors 11 monitoring the functioning of the production line 1 or the corresponding process step and a plurality of optical detection units 13, 13' (in particular cameras), of which at least one is arranged / arranged on the corresponding production line section 1.1 with a field of view 13a from above (in particular for anonymization) to view the corresponding production line section, preferably in an orientation at least approximately vertically downwards; wherein the system 100 is configured to transmit data (D) concerning at least one / the movement path 9 along at least two production line sections 1.1 of at least one specialist 7 supervising the production line 1 (or even a part of the specialist's body, e.g.to detect (of an arm) and, in response to a stop, to correlate at least one temporally predefinable sequence of the movement path 9 of the specialist 7 with at least one specific type of stop, in particular at least one error characteristic f1, f2, for preferably error-specific generation of at least one action recommendation 5 and for outputting at least one action request 5a, e.g. in file form, preferably in video file form, and / or on at least one visual and / or acoustic display unit 15 of the system 100.

[0071] In other words, the invention relates to a system 100 for error monitoring when a production line 1 for packaging material 3 is stopped, for detecting errors or their causes, and for generating at least one recommendation for action. The production line 1 has at least one open production line section 1.1 with at least one access point for manual error correction by a skilled worker. The production line 1 is, for example, divided into at least two or three functionally distinct process steps 1a, 1b, 1c, ..., 1i (or sections of the line), particularly with corresponding temporally / spatially successive production line sections. The use of the designation 1i for an i-th process step clarifies that the functional subdivision of the overall process is production line-specific, particularly with regard to downstream processes following gluing, such as...Stacking, sorting, or material transfer are affected. The output advantageously includes both action recommendations 5 and action requests 5a, whereby action recommendations 5 can also be generated, for example, within the framework of or with regard to documentation and training materials, and whereby action requests 5a are advantageously generated at least in near real-time directly in the field, possibly also in conjunction with the action recommendations 5. In particular, the action request 5a takes into account the current movement path 9 of a specialist (or a body part such as an arm of the specialist) involved in troubleshooting in response to a stop and is individually adapted by the system in this respect, e.g., with regard to the direction or type of recommended troubleshooting measure.

[0072] The respective field of view can be designed as a 360° all-round field of view or as a slightly oriented, in particular conical, field of view in or against the conveying direction (x), especially with a predefined oriented viewing axis, e.g., with an inclination angle in the range of 10 to 50° in or against the conveying direction. Advantageously, at least one detection unit with a field of view slightly inclined against the conveying direction is implemented, especially at the entrance in the area of ​​a material feed.

[0073] To adjust the advantageous field of view of a respective detection unit 13, 13', a holder 14 can be provided, in particular with at least one alignment component 14.1 (e.g., comprising several rods that can be aligned relative to each other, e.g., by means of clamping or friction-fit connections) and with at least one fastening component 14.3 (e.g., comprising a clamping lever or a clip connection). The holder can also be provided with an electrical supply and optionally a wired data connection to the production line.

[0074] It is understood that, for the sake of simplicity, each bracket 14 provides a static mount or a predefined orientation for the respective detection unit. In this respect, the desired level of detail in motion path detection can be achieved by selecting the appropriate number, arrangement, and orientation of the detection units. Alternatively, at least one of the brackets can also be equipped with at least one actuator for advantageously fully automatic camera tracking in the context of motion detection, for example, when the use of multiple brackets in a specific area or section of the production line is not particularly easy to implement. Therefore, the equipment complexity can also be reduced by using at least one camera tracking system, i.e., by using at least one movable detection unit.However, in terms of robustness and ease of implementation, it may be preferable to implement permanently installed, immobile acquisition units with an unchanging, predefined field of view, which can also facilitate or accelerate the analysis / evaluation of image data and irregularities.

[0075] Advantageously, two different types of optical detection units are provided, in particular a first type of detection unit 13 with a 360° field of view and a second type of detection unit 13' with a conical field of view oriented in or against the direction of conveying, whereby the width or angle of the field of view can be predetermined depending on the installation position and the area to be monitored. The first type of detection unit 13 with a 360° field of view is particularly well suited to detecting (macroscopic) movement paths of the respective specialist along the production line to the individual production line sections, while the second type of detection unit 13' with a field of view restricted in width or angle is particularly well suited to detecting shorter movement paths at the respective manual intervention point or rather small (microscopic) movements, e.g., only of body parts (hand, finger, arm), of the respective specialist.

[0076] The invention also relates to a computer program product 20 comprising commands which, when the computer program product 20 is executed on a computer, cause the computer to perform steps for controlling / regulating a process according to the present disclosure on the computer, in particular a computer program product configured to specify the manner of outputting at least one action recommendation 5 when monitoring at least one type of stop of the openly accessible production line 1 described herein for at least partially automatic processing of packaging material 3, in particular configured to specify the manner of recording physical movements of at least one skilled worker 7 and to specify time points T1,T2 and / or positions P2 when generating and outputting at least one action recommendation 5 and / or action request 5a for troubleshooting, particularly in response to and in temporal relation to a stop.

[0077] Output can occur at different locations and on different devices, e.g., on an end device 30 and / or on a production line-specific or, if necessary, external operator terminal 40. Output, particularly of instructions (especially output occurring at least approximately in real time), is advantageously performed at least at one output position P2. The computer program product 20 can be configured to provide a graphical user interface 41 with an input mask, particularly for requesting recommendations / instructions for action, whereby (e.g., password-protected) access to an administrator platform 70 can also be granted via the graphical user interface.

[0078] A corresponding control / regulation specification can optionally be generated via a central control / regulation system 50 or via a decentralized (especially production line-specific) control / regulation device 60. The system described here, or the corresponding production line, can be equipped with at least one communication module for at least one communication protocol.

[0079] Furthermore, an implementation is provided for a method for monitoring at least one type of stop of an openly accessible production line 1 for at least partially automatic, preferably fully automatic processing of packaging material 3, in particular cardboard boxes, for detecting the errors causing at least one type of stop, and for preferably generating (step S3) an error-specific recommendation for action 5 for troubleshooting, in particular for at least one specialist 7 operating / monitoring the production line 1, wherein a plurality of sensors 11 monitoring the functioning of the production line 1 and a plurality of optical detection units 13 are in communication with a computing unit 10 with data storage, wherein the optical detection units 13 have a field of view 13a from above a corresponding production line section 1.1. Inspect data (D) concerning at least the movement path 9 along at least two production line sections 1.1 by at least one specialist 7 supervising production line 1 is recorded (step S1), and in response to a stop, at least one temporally predefinable sequence of the specialist's movement path is correlated with at least one specific type of stop (step S2), preferably for the error-specific generation of at least one action recommendation and for the output (step S4) of at least one action request 5a, in particular in file form and / or on at least one visual and / or acoustic display unit. The sequence of the steps of recording (step S1), correlating (step S2), generating (step S3), and outputting (step S4) can also be varied, or the steps can be superimposed.The system may include further downstream steps (steps S5), particularly based on recorded user interactions (movement and / or user input), and especially based on the recording and evaluation (step S5a) of actual troubleshooting measures taken in the field. The time-predefinable sequence of the movement path (or the corresponding recording, step S1) may be defined within a time window T1 of, for example, at least three seconds before and / or at least five seconds after the stop. The system may also define at least one playback time window T2 for displaying at least one action prompt.

[0080] The following section explains special features of the invention with reference to individual figures or embodiments.

[0081] In the Fig. Figure 1 shows a first embodiment of an implementation of the system 100 described herein on a production line 1, wherein the production line is configured, for example, as a folding carton gluing machine with a machine speed in the range of approximately 250 m / min and, depending on the definition, at least three to five functionally distinct sections or process steps. Based on the present disclosure, the person skilled in the art understands that production line variations can each be addressed and represented in the system 100 described herein, for example, with regard to the number / arrangement / orientation of the detection units 13 and / or display units 15 described herein and / or the control / regulation of the individual steps S1 to S5 and / or the design and nature of the action prompts 5. Fig. Figure 1A shows that a skilled worker (7) is currently at the infeed point at the head of the production line. If a fault occurs much further downstream in process step 1c, the recommended actions / requests can be issued directly in that area. Three functionally distinct process steps (1a, 1b, 1c) are indicated as examples: the corresponding production line sections (1.1) for process step 1a (infeed), for several process steps (1b) (folding), and for process step 1c (skewed gluing, optionally followed by a reject module). It should be understood that further process steps, such as gluing, changing direction, stopping, and output / ejection, may be included, but these are not explicitly described here for the sake of clarity.It can be assumed that at least three consecutive or functionally distinct process steps are implemented, each of which must be able to effectively / efficiently rectify a potential cause of an error. The [details omitted] Fig. The sensors 11 indicated in Figure 1A for monitoring the functionality of the corresponding process step suggest that the implementation of the system 100 described here can be based not only on data collected by means of the acquisition units 13, 13a, but also on (measurement) data collected by means of additional sensors, in particular by means of production line-specific sensors that can be implemented individually for a respective process step, e.g., for feeding, folding, skewing, gluing, changing direction, stopping, and / or dispensing or ejecting. Since such sensors are advantageously implemented depending on the specific design of a given production line, a description of this remains general at this point.

[0082] In Fig. 1B shows the individually dimensioned, shaped and aligned viewing fields of the individual detection units. The two types of detection units 13, 13' are each positioned on brackets 14 and specifically aligned with their respective viewing fields 13a, in particular with four detection units with a viewing field aligned in the conveying direction (y) and one detection unit with a viewing field aligned against the conveying direction (y). Fig. 1B also indicates the longitudinal direction x of the production line (here corresponding to the conveying direction), the transverse direction y, and the vertical direction or axis z. It is understood that the skilled worker 7, shown on the far right of the diagram, would need a comparatively long time to search for the causes of errors along production line 1, simply in terms of the distance traveled.The system described here, advantageously at least in near real-time and also with AI support, can significantly facilitate and accelerate the analysis of potential errors, particularly in the context of machine stoppages (or for the preventive elimination of minor causes of errors that do not themselves result in an immediate machine stoppage), and also decouple it from the size-related disadvantages of a corresponding production line, in particular by indicating the respective location (position, spot) and advantageously also the diagnosed type of cause of the error in an acoustic and / or visual manner.For output (step S4), comparatively large display units 15 positioned at several points along the production line can also be used. This facilitates error detection (understanding the cause of an error) by the respective specialist involved, even from a great distance, and promotes the time-efficient and effective initiation of corrective measures, especially manual ones. Last but not least, this can also reduce the personnel required for a given production line, for example, by assigning responsibility for several production lines to a single specialist, or by having a production line previously supervised by several specialists now supervised by only one specialist.

[0083] In the exemplary embodiment according to the Fig. Figure 1 shows two detection units 13 of the first type (all-round field of view) and five detection units 13' of the second type (spatially oriented and optionally focused / restricted field of view) implemented. Based on the present disclosure, the person skilled in the art can also vary the type selection, number, position and orientation of the detection units and, for example, adapt them to specific batches.

[0084] In the Fig. Figure 2 shows an exemplary embodiment of the implementation of the system 100 described herein on an openly accessible packaging material production line 1 (e.g., a folding carton gluing machine). Based on the exemplary described / illustrated detection units and their number, arrangement, and orientation (especially their respective fields of view, illustrated here against the conveying direction (x)), a person skilled in the art will understand how the system 100 described herein can be implemented, particularly in conjunction with production line-specific sensor detection.

[0085] In Fig. Figure 2A shows a perspective view of a production line with numerous openly accessible sections or a multitude of successive, functionally distinct process steps. It is evident that the individual sections or process steps can also be coupled modularly, possibly in a different sequence than shown; thus, it becomes clear that processing individual batches of packaging material requires a high degree of process flexibility and variability, which can be ensured thanks to the design of the system 100 according to the invention in the context of the monitoring described here (see also [reference to be added]). Fig. 11) For the sake of completeness, it should be mentioned that in Fig. 2A the section which is described in detail in Fig. The 2D representation is not explicitly marked as a functionally distinct process section of the production lines; nevertheless, this section may also have its own function depending on the definition of its scope. Fig. Figure 2B shows the arrangement and orientation of a first detection unit of the second type 13' with a field of view against the conveying direction (x), particularly concerning the material intake. Based on this orientation and this restricted or focused field of view, the movement path of only one body part of the respective specialist (here also referred to as "microscopic") can be recorded particularly effectively. Fig. Figure 2C shows the arrangement and orientation of two detection units 13 of the first type with a round-the-round field of view, with the corresponding positions of these detection units 13 shown as in Fig. 2A shown, reference is made (in Fig. 2C these positions are summarized in a detailed representation). Using these all-round view detection units 13, a macroscopic movement path of the respective specialist can be recorded particularly effectively. In Fig. Figure 2D shows the arrangement and orientation of a second detection unit of the second type 13' with a field of view in the conveying direction (x), particularly concerning a stop operation with respect to the batch (material, carton), i.e., concerning an operation downstream of the production line section or process 1b, whereby the field of view is not necessarily conical, but can also be fan-shaped or restricted. The specific limitation of the field of view, especially depending on the monitored point or the recorded processing step, can also minimize the amount of data to be analyzed and make the evaluation more reliable and of higher quality. Fig. Figure 2E shows the arrangement and orientation of a third detection unit of the second type 13' with a field of view in the conveying direction (x), in particular with regard to a skewed gluing (production line section or process 1c). Fig. 2F shows the arrangement and orientation of a fourth and fifth detection unit of the second type 13' with a field of view in the conveying direction (x), in particular with regard to processing operations downstream of the (skew) gluing, in particular with regard to material dispensing, stacking, sorting or similar processes.

[0086] In the Fig. Figure 3 shows a perspective view illustrating an embodiment of the system 100 described herein on an openly accessible packaging material production line 1 (e.g., a folding carton gluing machine). Based on the exemplary described / illustrated detection units and their number, arrangement, and orientation (especially their respective fields of view, now illustrated in the conveying direction (x)), a person skilled in the art will understand how the system 100 described herein can be implemented, particularly in conjunction with production line-specific sensory detection.

[0087] In Fig. 3A is, in perspective view, a / the production line according to Fig. 2A shown, from a different perspective. In Fig. Figure 3B shows the arrangement and orientation of the first detection unit of the second type 13' with a field of view against the conveying direction (x). Fig. Figure 3C shows the arrangement and orientation of a first-type detection unit 13 with a 360° field of view, particularly in the process area of ​​folding the packaging material. Fig. The 3D diagram shows the arrangement and orientation of a second type 13' detection unit with a field of view in the conveying direction (x), particularly during the process of changing direction and / or stopping. Fig. Figure 3E shows the arrangement and orientation of a third detection unit of the second type 13' and a second detection unit of the first type 13, particularly in the procedural area of ​​a skewed adhesive bond. In Fig. 3F are the arrangement and orientation of a fourth and fifth detection unit, each of the second type 13' with field of view (and according to Fig. 3F (also viewing direction) shown in the direction of conveyance (x), particularly concerning material dispensing, stacking, sorting or similar processes.

[0088] In the Fig. 4. The fields of view of the respective detection unit are highlighted more strongly, each illustrated in the conveying direction (x), with reference to the further figures for components of the production line and / or the system, in particular to the Fig. 3A to 3F.

[0089] In Fig. 4A is a / the production line according to Fig. 3A is shown schematically, with the contrasting highlighted detection units 13, 13' and their fields of view 13a. In the Fig. 4B, Fig. 4C, Fig. 4D, Fig. 4E, Fig. 4F are / will be the views according to Fig. Figures 3B to 3F are repeated, with the acquisition units highlighted in contrast to the schematically depicted production line sections. Optionally, the production line can also be equipped with acquisition units at further sections or locations and in other orientations. For example, the respective acquisition unit is in data communication communication with a decentralized control / regulation device 60 and / or a central control / regulation system 50, as shown in Fig. 9 indicated.

[0090] In the Fig. 3 and Fig. For better understanding, the predefined oriented sight lines X13 of some of the detection units 13' focused in the field of view are also indicated by dashed lines. Due to the fact that in the Fig. The representation of the 3 superimposed components also refers to the relative arrangement and orientation of the respective detection unit. Fig. 4 indicated or referred to. In that respect, the one in the Fig. 3 and Fig. The revelation shown in section 4 can also be understood as a complementary revelation.

[0091] It is understandable that from those in the Fig. Of the display units shown in 1 to 4, at least one display unit may be designed or provided as a component of an operator terminal.

[0092] In the Fig. Figure 5 illustrates details of an exemplary embodiment of a bracket 14 for the respective detection unit 13, 13', wherein the respective bracket 14 can advantageously be adjusted both vertically and about at least one spatial axis, and advantageously about at least two spatial axes, to adjust the respective advantageous field of view of a respective detection unit 13, 13'. Advantageously, a first alignment component 14.1 comprises several rods that can be aligned relative to one another, e.g., by means of clamping or friction-fit connections, and by means of at least one fastening component 14.3 (e.g., a clamping lever or clip connection), e.g., a camera housing 13.1 can be attached to one of the alignment components 14.1. For the sake of completeness, it should be mentioned that the detection unit does not necessarily have to be formed by a camera.

[0093] In Fig. Figure 5A shows a rod-shaped support 14, here using the example of at least three rods and at least two adjustment axes. The support 14 itself is advantageously attached to a component of the production line, e.g. by means of the Fig. 5A at the lower end of the bracket 14 indicated plate, e.g. on a (machine) housing in the corresponding section of the production line. In Fig. Figure 5B shows the camera mounting or camera attachment, advantageously carried out at a free end of one / the last of the rods of the bracket.

[0094] In the Fig. Figure 6 illustrates an example situation for potential movement paths of several skilled workers along a production line. Fig. Figure 6A clearly illustrates the different ways a skilled worker can access one of several manual intervention points on the production line. Fig. As shown in section 6B, if a specialist makes an unfavorable maneuver in the wrong direction, a considerable amount of time may elapse before the specialist corrects the maneuver and returns to the correct course towards the source of the error. The system described here can effectively assist in this regard, for example, by generating acoustic outputs in addition to visual outputs in the local area of ​​the source of the error.

[0095] In the Fig. 7 will be those already in the Fig. The two components shown are taken up and presented in the context of different potential movement paths of at least one specialist.

[0096] In Fig. 7A indicates the direction (movement path 9) in which the respective specialist 7 can initiate a potential corrective action. Fig. Section 7B represents the first functionally distinct section of the production line. Fig. 7C is a detection unit with all-round vision shown. Fig. 7D shows a section in the area or after a stoppage of the material. In the Fig. 7E, Fig. Figure 7F shows a section after a skewed gluing. Fig. 7G is a section corresponding to a finishing step such as material removal.

[0097] In Fig. Figure 8 describes the steps of a procedure for monitoring at least one type of stop on a production line described herein. The procedure is implemented for detecting (steps S1, S2) the errors that cause at least one type of stop and for generating (step S3) at least one visually and / or audibly detectable recommendation for corrective action. At least one (momentary) movement path of at least one production line attendant is recorded (step S1), and in response to a stop, at least one temporally predefined sequence of the attendant's movement path is correlated with at least one specific type of stop (step S2). This allows for the preferably error-specific generation of at least one recommendation for action and the output (step S4) of at least one action request. The output can also occur at predefined locations along the production line, e.g.,The process takes place directly in the area of ​​the fault's cause (provided a display unit is located there). The sequence of the steps—capture (step S1), correlation (step S2), generation (step S3), and output (step S4)—can be varied, or the steps can be overlapped. The system may also include further downstream steps (step S5), particularly based on capturing and evaluating (step S5a) actual corrective actions taken in the field. This can be especially useful, for example, in the context of generating training materials (general recommendations for action). Time windows T1 for capture and / or T2 for output can also be defined.

[0098] The in Fig. The eight suggested diamond-shaped fields between the individual steps illustrate, for the expert, control / regulation measures that can be implemented in a production line-specific manner, particularly as a function of currently collected sensor data, for example also with regard to reactivating the production process after a stop.

[0099] In Fig. Section 9 describes the invention with reference to a schematically indicated production line, including further data processing components operating in conjunction with the computing unit described above. These data processing components may also be provided in the production lines or systems described above. The production line 1 comprises, in particular, an operator terminal 40, which can provide access to an administrator platform 70. A graphical user interface 41 with an input mask, particularly for requesting recommendations or instructions, can be operated or used via the computer program product 20 described herein. A central control system 50 can also be implemented at the operator terminal 40, optionally in addition to a decentralized control device 60.In particular, the data processing components explicitly mentioned here can each be wirelessly connected to each other via communication modules 80. Advantageously, the operator terminal 40 is connected to the respective production line 1, whereas the administrator platform 70 can optionally be implemented in a way that allows access from other locations or data processing components, for example, via the (decentralized) control unit 60. Based on the present disclosure, a person skilled in the art can specify at which locations and via which components access authorization may be implemented. For example, it may be useful to enable the respective specialist to issue recommendations / requests for action regardless of location, for example, to repeatedly play a specific video sequence, especially in time-lapse, whereas, particularly for safety reasons, it may be advantageous to allow a restart or...To enable a restart of the processing process exclusively via the administrator platform 70, specifically only from a single position along the production line (advantageously at the operator terminal 40 as the main instance). Alternatively, the operator terminal 40 can be a single operator terminal 40 operating jointly for multiple production lines. Fig. Figure 9 indicates that the respective terminal device 30 may also be equipped with a visual and / or acoustic display unit 15. The specialist 7 or the involved user can optionally receive a recommendation / request for action on at least one stationary display unit and / or on a terminal device.

[0100] In Fig. Figure 9 depicts a situation in which a fault cause has been diagnosed in the area of ​​process step 1b, wherein the corresponding (at least one triggered or controlled) display unit 15 is advantageously also arranged in this production line section 1b, 1.1; the user 7 can then reach the corresponding playback position P2 or the location of the fault cause in a time-efficient manner, in particular according to movement path 9.

[0101] It is understood that the relative arrangement of detection units, display units, sensors, and similar components for data processing shown in the figures can be varied; in particular, the respective position of a detection unit with respect to the position of sensors and / or open sections with manual intervention options can be varied. For example, components 10, 40, 50, and 60 shown in the figures can also be positioned variably, especially since the respective production line can be assembled modularly for the individual, functionally distinct process steps, so that optionally a separate or specific control system may be provided for each process step. The [unclear text] Fig.The Terminal 40 shown in Figure 9 may be positioned differently, for example in such a way that it is at least visually accessible via a central display unit for several production lines within a production hall or machine hall. Reference symbol list 1 production line 1.1 open production line section 1a, 1b, 1c, ..., 1i functionally distinct process steps or sections 3 Packaging material 5, 5a Recommendation for action, call to action 7 skilled worker 9. Career path of a specialist 10 computing unit 11 Sensor 13 optical detection units of the first type (especially with all-round field of view) 13' optical detection unit of the second type (especially with focused field of view) 13a Field of vision 13.1 Camera body 14 Bracket for detection unit 14.1 Alignment component of the bracket 14.3 Mounting component of the bracket 15 Display unit 20 computer program product 30 terminals 40 operator terminals 41 graphical user interface with input mask 50 central control / regulation system 60 Control / regulating device 70 Administrator Platform 80 Communication module D captured (image / video) data P2 at least one playback position for output S1 step of recording S2 step of correlation S3 step of generation S4 step of output S5 downstream steps, particularly those based on user interaction S5a Recording and evaluating corrective actions T1 Time window for recording movement / movement path of the relevant specialist T2 at least one playback time window for output X13 predefined aligned line of sight x, y, z Longitudinal direction (or conveying direction), transverse direction, vertical direction QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] DE 603 ​​17 073 T2

[0004]

Claims

[1] System for monitoring at least one type of stop of a production line for at least partially automatic, preferably fully automatic processing of packaging material, in particular cardboard boxes, for detecting errors / causes causing the at least one type of stop, and for preferably error-specific generation of at least one visually and / or acoustically detectable recommendation for action to rectify the error, in particular for at least one specialist operating / monitoring the production line, wherein the respective production line is an openly accessible production line of e.g.over 10m in length with several functionally separable process steps, each with manual intervention option for manual error correction in the corresponding area / production line section of the respective process step, wherein the system has a computing unit with data memory which is in communication with a plurality of sensors monitoring the functioning of the production line and a plurality of optical detection units, of which at least one can be arranged / is arranged on the corresponding production line section with a field of view from above, preferably in an orientation at least approximately vertically downwards; . characterized bythat the system is set up to record data relating to at least the movement path along at least two production line sections of at least one specialist supervising the production line and, in response to a stop, to correlate at least one temporally pre-definable sequence of the movement path of the specialist with at least one specific type of stop, for preferably error-specific generation of the at least one recommended action and for outputting at least one request for action, e.g. in file form, preferably in video file form, and / or on at least one visual and / or acoustic display unit of the system. [2] System according to claim 1, wherein the system is configured, in response to a stop, to correlate the last acquired data relating to at least the movement path or the temporally pre-definable sequence of the movement path before / after the stop with at least one type of stop and to generate a request for action in the form of at least one error correction measure and to output it on at least one visual and / or acoustic display unit of the system, in particular within five to ten seconds at the latest after the stop has occurred, advantageously with the visual and / or acoustic output being predetermined locally in the area of ​​at least one affected production line section. [3] System according to one of the preceding claims, wherein the temporally predefinable sequence of the movement path is defined to a time window of at least three seconds before the stop and / or at least five seconds after the stop. [4] System according to one of the preceding claims, wherein at least one playback time window is defined on the system side for outputting the at least one call to action, in particular at least a first playback time window for outputting a first call to action and a second playback time window for outputting a second call to action. [5] System according to one of the preceding claims, wherein at least one playback position for outputting the at least one request for action is / is defined on the system side, in particular at least a first playback position for outputting a / the first request for action and a second playback position for outputting a / the second request for action. [6] System according to one of the preceding claims, wherein the system is configured to perform the output based on a time-lapse, in particular by outputting the data last acquired in response to a stop relating to at least the movement path or the corresponding temporally predefined sequence of the movement path as a video at a speed reduced by at least a factor of two, in particular on at least one of the display units. [7] System according to one of the preceding claims, wherein the system is configured to generate the at least one recommended action at least partially based on data stored in the data memory, in particular after correlation has taken place. [8] System according to one of the preceding claims, wherein at least one specific type of stop is / is correlated with at least one defect feature (f1, f2, ...), in particular at least one defect feature from the following group: defect position, defect time, batch identification feature, production line identification feature, functional processing step, spatial orientation of the workpiece; in particular specifically with reference to a specific production line and / or a specific packaging material batch. [9] System according to one of the preceding claims, wherein the system is arranged to carry out the acquisition of the data after reaching a pre-definable target machine performance or target production line performance, in particular in a continuous manner by only temporarily storing a pre-definable past time window of the acquired data and only saving it in the event of a stop or storing it permanently in the data memory. [10] System according to one of the preceding claims, wherein the system is configured to carry out at least the detection and, depending on the stop, also a correlation based on at least one performance / output parameter, in particular depending on a fluctuation of the at least one performance / output parameter, in particular when a pre-definable deviation from a target performance / output parameter is exceeded. [11] System according to one of the preceding claims, wherein the system is configured to perform at least the correlating and / or at least the detecting by means of at least one or based on at least one ML algorithm. [12] System according to one of the preceding claims, wherein at least one optical detection unit is arranged on each openly accessible production line section with manual intervention for manual error correction, preferably aligned at least approximately vertically downwards, for example at a height of three or four meters. [13] A computer program product comprising instructions which, when executed on a computer, cause the computer to carry out steps for controlling / regulating the following method on the computer: a method for monitoring at least one type of stop of an openly accessible production line for at least partially automatic, preferably fully automatic processing of packaging material, in particular cardboard boxes, for detecting errors / causes causing the at least one type of stop, and for preferably error-specific generation of at least one visually and / or acoustically detectable recommended action for troubleshooting, in particular for at least one specialist operating / monitoring the production line, wherein a plurality of sensors monitoring the functioning of the production line and a plurality of optical detection units are in communication with a computing unit with data memory,wherein the optical detection units view a corresponding production line section with a field of view from above; wherein data relating to at least the movement path along at least two production line sections is recorded by at least one specialist supervising the production line, and in response to a stop, at least one temporally predefinable sequence of the specialist's movement path is correlated with at least one specific type of stop, for the preferably error-specific generation of the at least one recommended action and for outputting at least one request for action in file form and / or on at least one visual and / or acoustic display unit. [14] Computer program product according to the preceding claim, wherein the computer program product is configured for a control / regulation in which, in response to a stop, the last acquired data relating to at least the movement path or the temporally pre-definable sequence of the movement path before / after the stop are stored and correlated with at least one type of stop and a request for action in the form of at least one error correction measure is generated and output on at least one visual and / or acoustic display unit or the corresponding production line. [15] Computer program product according to one of claims 13 to 14, the computer program product is set up for a control / regulation in which the temporally predefinable sequence of the movement path is specified for a time window of at least three seconds before the stop and / or at least five seconds after the stop. [16] Computer program product according to one of claims 13 to 15, wherein the computer program product is configured for a control / regulation in which at least one playback time window is specified for outputting the at least one request for action, in particular at least a first playback time window for outputting a first request for action and a second playback time window for outputting a second request for action. [17] Computer program product according to one of claims 13 to 16, wherein the computer program product is configured for a control / regulation in which at least one playback position is specified for outputting the at least one request for action, in particular at least a first playback position for outputting a / the first request for action and a second playback position for outputting a / the second request for action. [18] Computer program product according to one of claims 13 to 17, wherein the computer program product is configured for control / regulation in which the output is based on a time-lapse, in particular by outputting the data last acquired in response to a stop relating to at least the movement path or the corresponding temporally predefined sequence of the movement path as a video at a speed reduced by at least a factor of two. [19] Computer program product according to one of claims 13 to 18, wherein the computer program product is configured for a control / regulation in which the generation of the at least one recommended action is carried out at least partially based on data stored in the data memory, in particular after correlation has taken place. [20] Computer program product according to one of claims 13 to 19, wherein the computer program product is configured for a control / regulation in which the generation of the at least one recommended action comprises the generation of an error description. [21] Computer program product according to one of claims 13 to 20, wherein the computer program product is set up for a control / regulation in which the acquisition of the data takes place after reaching a pre-definable target machine performance or target production line performance, in particular in a continuous manner by only temporarily storing a pre-definable past time window of the acquired data and only storing it in the event of a stop or permanently storing it in the data memory. [22] Computer program product according to one of claims 13 to 21, wherein the computer program product is configured for a control / regulation in which at least the detection and, depending on the stop, also a correlation is carried out based on at least one performance / output parameter, in particular depending on a fluctuation of the at least one performance / output parameter, in particular when a predefinable deviation from a target performance / output parameter is exceeded. [23] Computer program product according to one of claims 13 to 22, wherein the computer program product is configured for a control / regulation in which at least the correlation is carried out by means of at least one or based on at least one ML algorithm, in particular based on at least one AI model implemented for the stop-dependent error detection. [24] Computer program product according to one of claims 13 to 23, wherein the computer program product is configured for control / regulation in which at least the detection is carried out by means of at least one or based on at least one ML algorithm, in particular based on at least one AI model implemented for detecting a movement of a specialist, in particular for evaluating a / the current movement of a specialist in the context of an error-specific recommendation / request for action. [25] A computer program product comprising instructions which, when executed on a computer, cause the computer to carry out steps for controlling / regulating a method for monitoring at least one type of stop of an openly accessible production line for at least partially automatic processing of packaging material for detecting errors / causes causing the at least one type of stop on the computer, and for generating on the computer, in a fault-specific manner, at least one visually and / or acoustically detectable recommended action for troubleshooting, in particular for at least one specialist operating / monitoring the production line, in which method a plurality of sensors monitoring the functioning of the production line and a plurality of optical detection units are in communication with a computing unit with data memory,wherein the optical detection units view a corresponding production line section with a field of view from above; in which method, data relating to at least the movement path along at least two production line sections are recorded by at least one skilled worker supervising the production line, and in response to a stop, at least one temporally predefinable sequence of the movement path of the skilled worker is correlated with at least one specific type of stop,for the preferably error-specific generation of at least one recommended action and for outputting at least one request for action in file form and / or on at least one visual and / or acoustic display unit; in particular, a computer program product configured to specify the manner of outputting at least one recommended action when monitoring at least one type of stop of an openly accessible production line for at least partially automatic processing of packaging material, in particular configured to specify the manner of detecting physical movements of at least one skilled worker and to specify times and / or positions when generating and outputting at least one recommended action and / or request for action for error correction, in particular in response to and in temporal connection with a stop. [26] Use of a system in communication with a plurality of sensors monitoring the functioning of an openly accessible packaging material production line and with a plurality of optical detection units, comprising a computing unit and data storage and a communication module for monitoring at least one type of stop of the production line, for detecting errors / causes causing the at least one type of stop, and for generating at least one visually and / or acoustically detectable recommended action for troubleshooting for at least one specialist supervising the production line, wherein the optical detection units are / are arranged along the production line so as to be visible to a corresponding production line section, wherein data relating to at least the movement path along at least two production line sections is recorded by at least one specialist,wherein, in response to a stop, at least one temporally pre-definable sequence of the movement path of the corresponding specialist, in particular a course over several seconds immediately after the stop, is correlated with at least one specific type of stop, in order to generate the at least one recommended action and to issue at least one request for action. [27] Use of a computer program product for specifying the manner of recording and evaluating (video) data relating to processing processes of a production line for processing packaging material, and for implementing an error analysis and error correction instruction functionality in coordination with a central control / regulation system of the production line in response to a stop of the production line, and for outputting at least one request for action to correct the error, wherein the output further comprises at least outputting a video relating to a time window from a few seconds before the stop to a few seconds after the stop,in particular a video comprising both image data relating to the course of the origin of the cause of the error or the temporal course of the machining at the error location as well as image data relating to the movement path or a / the temporally pre-definable sequence of the movement path of the corresponding specialist, in particular immediately after the stop. [28] Production line (1) for at least partially automatic processing of packaging material (3), comprising a system (100) according to one of claims 1 to 12.

Images