DETECTION OF RECIRCULATING OBJECTS

Description

[0001] The invention relates to the automation of logistics systems for piece goods with return.

[0002] In sorting systems for parcels or suitcases with automated unloading and singulation, the objects to be sorted are transported as 2D / 3D bulk material on conveyor belts to a singulation system (e.g., Visicon, robot, Variotip). An oriented 1D stream with defined gaps between the objects is then generated.

[0003] The objects to be sorted are general cargo, such as parcels, shipping goods, or even luggage. Typical applications can be found in sorting systems at airports or at postal or other logistics service providers. The terms "shipment," "general cargo," and "object" are used essentially synonymously below.

[0004] Errors can arise due to deficiencies in the sorting and / or singulation system (e.g., in mechanical components, detection technology, or special feeding situations) or in certain object types. These errors can include, for example, double feeds, incorrect orientation of a piece of goods, or an insufficient gap between the next object.

[0005] An automatic detection system detects these errors and can recirculate the affected shipments and feed them back into the singulating system via a recirculation conveyor line (circular line, loop).

[0006] WO 99 / 21663 A1 shows a logistics system for piece goods, which separates piece goods and feeds them to a detection system.

[0007] In addition, there is the option of removing shipments completely from the stream, i.e. sending them for manual checking and further processing.

[0008] Certain objects are repeatedly recirculated. For example, the cause may be misidentification, where a single object is incorrectly identified as a duplicate. Such misidentifications can also be due to defective objects, open shipments (e.g., open packages or suitcases), irregularly shaped objects (e.g., L-shaped boxes), special prints or stickers on the objects, protruding stickers or strips of adhesive tape, or glued objects that do not separate even upon re-singling.

[0009] A similar problem arises if an address, barcode, or label cannot be read in an already singulated stream. For example, in a baggage sorting system such as that used in airports, at a reading station that reads a barcode on a baggage tag attached to a piece of luggage to sort baggage, an error in the reading process can be diverted to a return route that brings the baggage back to the reading station. However, if the tag is damaged and the barcode is illegible, the error is not corrected, so the baggage in question continues to circulate until the error is corrected, for example by manually handling the baggage.

[0010] During operation, the rate of these recirculating objects increases, and they are detected by the detection system at periodic intervals depending on the round-trip time (RTT). This results in increased mechanical stress on both the objects and the system, as well as a reduction in system throughput. This usually requires manual intervention, i.e., manual removal of suspicious objects, manual emptying of the system, or the problem and associated disadvantages must be accepted as long as the sorting system is not severely impaired and remains functional.

[0011] The present invention is based on the object of optimizing the operation of a logistics system for piece goods with return.

[0012] This problem is solved by the concepts described in the independent claims.

[0013] According to one aspect, the invention relates to a logistics system for piece goods. The logistics system comprises a main conveyor line, a detection system, a discharge system, and a recirculation conveyor line. The main conveyor line is designed to feed piece goods to the detection system and to convey them further downstream from there. The discharge system is designed to feed piece goods that were fed to the detection system from the main conveyor line to the recirculation conveyor line. The recirculation conveyor line is designed to feed piece goods that were fed to it by the discharge system back to the detection system. The detection system is configured to generate a data set for a piece goods complex fed to it, by means of which data set the piece goods complex can be identified. A piece goods complex can comprise a single piece goods item or multiple piece goods items.For example, a piece goods complex can be or include a piece goods cluster that has not been correctly singulated. In another exemplary embodiment, a piece goods complex is a single, singulated piece goods. The detection system is further configured to detect an error affecting a piece goods complex fed to it and to cause the discharge system to feed the error-affected piece goods complex to the recirculation conveyor line. The detection system is further configured to store at least those data records by means of which piece goods complexes fed to the recirculation conveyor line can be identified. The detection system is further configured to compare data records with one another and, if two or more data records match, to classify this match as evidence of a piece goods complex repeatedly affected by an error.

[0014] According to one aspect, the invention relates to a method for providing data or for operating a logistics facility for piece goods. In this case, piece goods are conveyed in the logistics facility along a main conveyor line to an automatic detection system. The detection system automatically detects an error affecting a piece goods complex. The detection system automatically creates a first data set by means of which the piece goods complex can be identified. Based on the detected error, the piece goods complex is diverted from the main conveyor line to a recirculation conveyor line. The recirculation conveyor line causes the piece goods complex to be fed back to the detection system. When the piece goods complex is fed back in, the detection system again detects the error affecting the piece goods complex and creates a second data set by means of which the recirculated piece goods complex can be identified.In a further process step, the recognition system detects a match between the first and the second data set and, based on this, classifies the first and the second data set as identifying the same piece goods complex and the piece goods complex as repeatedly affected by errors.

[0015] Advantages and embodiments of the invention, which can be used individually or in combination with one another, are the subject of the subclaims.

[0016] According to one embodiment, the logistics system comprises a singulator. The singulator is designed to singulate unit loads fed to the detection system. The error that causes the detection system to direct the reject system to feed the faulty unit load complex to the recirculation conveyor line is an error in the singulation of the unit load complex. This creates the conditions for automatically detecting faulty unit loads.

[0017] According to one embodiment, the recognition system determines a selection of the following data that can be stored in the data set: Time of recognition; Image information of the piece goods, a part of the piece goods or a piece goods cluster; Extracted data of the piece goods, a part of the piece goods or a piece goods cluster, such as one or more heights, one or more volumes, one or more dimensions, one or more colors, one or more types, one or more image fingerprints, one or more surface textures; One or more barcodes attached to the piece goods or to the piece goods cluster; One or more plain texts attached to the piece goods or to the piece goods cluster and extracted by means of text recognition; One or more conveyor speeds.

[0018] According to one embodiment, the method step of detecting a match between the first and second data sets comprises comparing the second data set with a plurality of stored data sets.

[0019] According to one embodiment, the recognition system is configured to apply a preferably time-based search space restriction when comparing data sets. This enables efficient execution of data set comparisons.

[0020] According to one embodiment, the detection system is further configured to determine a probability for the match between two or more threshold values ​​and, if the probability exceeds a threshold value, to classify the match as evidence of a repeatedly faulty piece goods complex. This enables the method step of detecting a match between the first and second data sets to automatically determine a probability for the match and automatically interpret an exceedance of the threshold value as evidence of a repeatedly faulty piece goods complex. In this way, piece goods complexes can be classified as repeatedly faulty if the error has not been resolved after one or more recirculations but has nevertheless changed.

[0021] According to one embodiment, the logistics facility is a sorting facility. The data contained in a data set is also used to sort the piece goods assigned to the data set. This allows for synergies to be utilized by avoiding unnecessary redundancy in the complex analyses of a piece goods, as the data contained in a data set can also be used to sort the piece goods assigned to the data set.

[0022] According to one embodiment, the logistics system comprises a manual processing station. The detection system is configured to cause the rejection system to feed a multiply faulty piece goods complex to the manual processing station. This allows for automatic relief of the recirculation conveyor line by automatically feeding a piece goods complex classified as multiply faulty to a manual processing station.

[0023] Further features, characteristics, and advantages of the present invention will become apparent from the following description with reference to the accompanying figures, which schematically show: Figure 1 a block diagram of a logistics system according to an embodiment of the invention; Figure 2 a block diagram of a logistics system according to a further embodiment of the invention.

[0024] In the exemplary embodiments and figures, identical or similarly functioning elements may be provided with the same reference numerals. The illustrated elements and their relative sizes are generally not to be considered to scale; rather, individual elements may be shown larger in proportion for clarity and / or clarity.

[0025] Figure 1schematically shows a logistics system 1 for piece goods, which includes an automatic exception detection system 3 with a 3-way split 4 and recirculation 5. The logistics system 1 includes a main conveyor line 2, a detection system 3, a discharge system 4, a recirculation conveyor line 5, a singulation system 6, a manual processing station 7, a merging system 8, and a control system 9.

[0026] The control system 9 is designed and adapted to control the logistics system 1, or at least a selection of the systems 2, 3, 4, 5, 6, 7, 8.

[0027] At least the recognition system 3 comprises hardware and software or logic. The hardware comprises a camera system 39 or other imaging system with a monitoring area, as well as a control system 38. The software or logic can be implemented in the control system 38 and directly connected to the discharge system 4 in order to control it. Alternatively or additionally, at least part of the software or logic of the recognition system 4 can be implemented in the higher-level control system 9 of the logistics facility 1. In some embodiments, functionalities of the recognition system 3 are therefore implemented in the control system 9, as can be seen by the dashed block 3, which represents the recognition system 3 and which overlaps with the control system 9.

[0028] The main conveyor line 2 is designed to guide piece goods from a feed 21 to the merging system 8, from there to the singulation system 6, from there to the detection system 3, from there to the discharge system 4, and from there to possible further processing systems 22.

[0029] Upstream of the feed 2, feeding systems can be arranged that perform logistical processing steps for the piece goods. For example, the piece goods can be delivered in a random manner similar to bulk goods, and the upstream systems generate a single-layer piece goods flow from stacked piece goods, which is fed to the merging system 8. From the merging system 8, the piece goods are fed to the singulation system 6, where the single-layered supplied piece goods are singulated. In one variant, the piece goods flow with stacked piece goods is fed to the singulation system 6, and the singulation system 6 generates a single-layer singulated flow of piece goods from it.

[0030] The recognition system 3 comprises a camera system configured to generate digital images of the piece goods fed to the recognition system 3. The piece goods are optimally fed to the recognition system 3 singulated. However, if an error occurs in the singulation system 6, it may happen, for example, that a piece goods cluster 24 comprising two or more non-singulated piece goods is sent to the recognition system 3.

[0031] Initially, however, it is not known whether an error has occurred in the singulation or not, but first, the recognition system 3 is supplied with piece goods complexes 23, 24 from the singulation system 6, which can be a correctly singulated piece goods 23 or a faulty piece goods cluster 24. In Figure 1the reference number 23 generally designates general cargo complexes which comprise only one correctly separated piece of cargo, while the reference number 24 generally designates general cargo clusters which comprise several pieces of cargo.

[0032] The recognition system 3 is configured to use digital image processing to determine from one or more of the digital images whether a piece of cargo has been correctly singulated or whether it is part of a piece of cargo cluster. The recognition system 3 is thus configured to determine whether a piece of cargo complex 23, 24 is a correctly singulated piece of cargo 23 or a faulty piece of cargo cluster 24. The recognition system 3 is therefore tasked with monitoring the quality of the singulation.

[0033] In the Figure 1 In the embodiment shown, the discharge system 4 is designed as a 3-way splitter and is set up to separate a piece goods complex a) either to be conveyed further along the main conveyor line 2; or b) to be discharged from the main conveyor line 2 and fed to the recirculation conveyor line 5; or c) to be discharged from the main conveyor line and fed to the manual processing station 7.

[0034] In one variant, the discharge system can also comprise two 2-way splitters, wherein a first of the 2-way splitters is designed to discharge a piece goods complex from the main conveyor line to the recirculation conveyor line 5, while a second of the 2-way splitters is designed to feed a piece goods complex from the main conveyor line 2 to the manual processing station 7.

[0035] If the detection system 3 interprets a piece goods complex 23, 24 as correctly separated piece goods 23 and thus as not affected by an error, it controls the discharge system 4 so that it conveys the piece goods complex 23, or in other words this piece goods 23, further along the main conveyor line 2. However, if the detection system 4 detects that an incoming piece goods complex is a piece goods cluster 24 that comprises two or more non-singulated piece goods, the detection system interprets this piece goods complex 24 as incorrectly separated and thus as affected by an error and controls the discharge system 4 so that it feeds the piece goods cluster to the recirculation conveyor line 5.

[0036] The recognition system 3 is also configured to generate a data set 33, 34, 35, 36 for each delivery of a piece goods complex 24 identified as being affected by an error to the recognition system 4, by means of which the piece goods complex can be identified. A data set 33, 34, 35, 36 can, for example, comprise a fingerprint of the piece goods complex, which represents geometric, optical, and / or physical properties of the piece goods complex, for example in the form of a vector. Instead of a vector, the extracted data can also be concatenated into a number or a string, which then forms a fingerprint, e.g., length_width_height_color. A deep learning method can also be used to determine a value or character string from the image that cannot be directly assigned to human-understandable values ​​such as "length."

[0037] According to one embodiment, the recognition system is configured to generate a data record only for those piece goods complexes that have been identified as being affected by an error. According to an alternative embodiment, the recognition system is configured to also generate a data record for other or all piece goods complexes supplied to the recognition system.

[0038] The recognition system 3 is also configured to store at least those data records 33, 34, 35, 36 in a database 31 by means of which the piece goods complexes fed to the recirculation conveyor line 5 can be identified. Alternatively, data records of all piece goods complexes fed to the recognition system 3 can also be generated and stored.

[0039] The recirculation conveyor line 5 is configured to first re-feed a cluster of incorrectly singulated piece goods, which was fed to it by the discharge system 4, to the singulation system 6 via the consolidation system 8. In one variant, the logistics system does not comprise a separate consolidation system 8; instead, the recirculation conveyor line 5 is configured to re-feed a cluster of incorrectly singulated piece goods, which was fed to it by the discharge system 4, directly to the singulation system 6. For example, the recirculation conveyor line 5 can comprise one or more belt conveyors.

[0040] If the singulation system 6 correctly singulates the recirculated piece goods cluster in this pass, the correctly singulated piece goods are fed to the detection system 6. In this pass, the detection system 6 recognizes these piece goods complexes as correctly singulated piece goods and controls the discharge system 4 so that these singulated piece goods are fed further along the main conveyor line to the other processing systems 22.

[0041] If the singulation system 6 is also unable to singulate the recirculated general cargo cluster 24, for example because the general cargo is stuck together or tangled, the recognition system 3 will again recognize the general cargo complex as a general cargo cluster and thus interpret it as affected by an error and create a second data record 34 for this general cargo cluster, by means of which the general cargo complex 24 can be identified again. The recognition system compares this second data record 34 with the data records stored in the database 31.If the second data set 34 matches the first data set stored in the database 31, the recognition system 3 classifies the first and second data sets 31, 34 as identifying the same piece goods complex and the piece goods complex as repeatedly affected by errors, or as having already been recirculated at least once, and controls the discharge system 4 to feed this piece goods complex to the manual processing station 7. In this way, a piece goods cluster that the singulation system 6 is unable to singulate can be prevented from unnecessarily burdening the recirculation conveyor line 5. Instead of a match between two data sets, the recognition system 3 can also be configured to discharge the piece goods complex in question to the manual processing station 7 only when three or more data sets match.In this way, unnecessary strain on the manual processing station can be avoided if two passes through the singulation system are not sufficient to singulate a piece of goods cluster.

[0042] In the Figure 1 In the embodiment shown, the discharge system 4 is arranged downstream of the detection system 3. In other embodiments, the discharge system 4 and the detection system 3 can be arranged in a common section of the main conveyor line 2 or partially overlap. The camera system of the detection system 3 covers a monitoring area that also includes the discharge system 4.

[0043] In the Figure 1In the embodiment shown, the singulation system 6 is arranged upstream of the detection system 3. In other embodiments, the singulation system 6 and the detection system 3 can be arranged in a common section of the main conveyor line 2 or partially overlap. The camera system of the detection system 3 covers a monitoring area that also includes the singulation system 6.

[0044] In a further embodiment, the detection system 3 can also cover a monitoring area which covers both the singulation system 6 and the ejection system 4.

[0045] In the Figure 1In the embodiment shown, the error that affects a piece goods complex fed to the detection system 3 is a singulation error. In other embodiments, the error that affects a piece goods complex fed to the detection system 3 is not a singulation error, but a different error. Such an embodiment of a logistics system which, like the embodiment of Figure 1 can be used in a sorting system, is in Figure 2 shown schematically.

[0046] Figure 2shows a logistics system 101 for general cargo such as parcels or pieces of luggage. The sorting system 101 is constructed similarly to the logistics system 1, however, the logistics system 1 does not include a singulation system and a modified recognition system 103. In a baggage sorting system such as those used in airports, this is generally not necessary, since the pieces of luggage are already individually handed onto the sorting system by an operator or a passenger. Labels representing a sorting destination are affixed to or on the general cargo. In the case of parcels, the sorting destination can be, for example, an addressee or a sorting or distribution center to which the parcel is to be transported. In the case of pieces of luggage, the sorting destination can be, for example, a target or intermediate destination of a piece of luggage, such as a destination airport.The information required for this can be placed on the label, for example in the form of a barcode, in the form of plain text, or stored in another information carrier such as an RFID.

[0047] If the recognition system 103 fails to extract the information necessary for sorting, for example because a label is damaged or obscured, the respective piece of goods is recognized by the recognition system 3 as being affected by an error, and the piece of goods is diverted to the recirculation conveyor line 5. In this case, the error is therefore a sorting target that is at least not completely readable.

[0048] The recognition system 103 is configured to generate a data record for each piece of cargo by means of which the piece of cargo can be identified.

[0049] The recognition system 103 is also configured to read the destination on the labels and use it in the further processing systems 22, which may, for example, include a sorter for sorting the piece goods. If a sorting destination for a piece of goods is fully recognized by the recognition system, the recognition system 103 causes the discharge system to feed this piece of goods to the further processing systems 22.

[0050] However, if a sorting destination for a piece of goods is not fully recognized by the recognition system 103, the recognition system 103 causes the discharge system 4 to interpret this piece of goods as being affected by an error and to feed it to the recirculation conveyor line 5.

[0051] The recognition system 103 stores a data record for each piece of goods fed to the recirculation conveyor line 5, by means of which the piece of goods can be identified.

[0052] A piece of goods affected by an error and diverted into the recirculation conveyor line 5 is thus fed again to the recognition system 103. If the sorting destination is readable this time, for example because the label is no longer covered, the recognition system 103 causes the discharge system 4 to feed the piece of goods to the further processing systems 22. If the sorting destination is again unreadable, the recognition system again creates a second data set for this piece of goods, by means of which the piece of goods can be identified and compares the second data set with all data records in the database 31 and finds a match with the previously created first data set for this piece of goods. Therefore, the piece of goods is classified as repeatedly affected by an error and causes the discharge system 4 to feed the piece of goods to the manual processing station 7.

[0053] According to further embodiments, the recognition system 3 is configured to apply a preferably time-based search space restriction when comparing data records. The restriction can be applied, for example, based on the circulation time of a piece of goods around the recirculation conveyor line, with only data records created, for example, somewhat longer than one or more circulation times ago being compared. The search space restriction can also be timed based on circulation times, so that, for example, only data records created in a time interval covering one or more points in time that lie exactly one or more circulation times ago are compared.

[0054] The recognition system may not be able to recognize a complex of items with 100% certainty. Instead, it must calculate an overall probability that the item is a previously seen item based on various factors by weighting the individual matches (e.g., length 100% identical, width 90% identical, color identical, height 50% identical). Accordingly, a threshold (e.g., 80% certainty) must be used to determine the next action.

[0055] According to further embodiments, the logistics system is a sorting system, and the data contained in a data set is additionally used for sorting the piece goods assigned to the data set. For this purpose, the recognition system can be connected to other systems of the sorting system, for example a sorter. The recognition system 3, 103 is configured to transmit data or data sets to the sorter or the other systems of the sorting system, which data or data sets comprise, for example, a fingerprint or a barcode or other identification by means of which the piece goods can be identified or sorted. In this way, synergies can be utilized for error detection and sorting. For example, this can avoid time-consuming, unnecessary multiple fingerprinting processes.

[0056] According to further embodiments, a data set comprises a selection of the following data: Time of recognition; Image information of the piece goods, a part of the piece goods or a piece goods cluster; Extracted data of the piece goods, a part of the piece goods or a piece goods cluster, such as one or more heights, one or more volumes, one or more dimensions, one or more colors, one or more types, one or more image fingerprints, one or more surface textures; One or more barcodes attached to the piece goods or to the piece goods cluster; One or more plain texts attached to the piece goods or to the piece goods cluster and extracted by means of text recognition; One or more conveyor speeds.

[0057] According to further embodiments, the creation of image fingerprints is trained into the recognition system using deep learning methods.

[0058] According to one embodiment, the discharge system is configured to selectively feed piece goods that have been fed to the detection system 3 to a manual processing station 7. The detection system 3 is configured to cause the discharge system 4 to feed a piece goods complex affected by multiple errors to the manual processing station 7.

[0059] According to further embodiments, a data set comprises an identifier by means of which the piece goods complex can be determined on the basis of which the recognition system created the data set.

[0060] According to further embodiments, the logistics system 1, 101 is configured to track a recirculated piece goods complex along the recirculation conveyor line 5. This can be implemented, for example, such that the monitoring area of ​​the detection system 3, 103 covers the entire recirculation conveyor line.

[0061] Further embodiments include the use of various process-based methods: According to one embodiment, the recirculation conveyor line 5 can be regularly emptied. For a circulation period RTT, all detected faulty piece goods complexes are automatically conveyed from the recirculation conveyor line 5 to the manual processing station. This reduces the effective detection rate and increases personnel costs. If the supply of new piece goods complexes is additionally stopped, the throughput decreases instead, without adversely affecting the detection rate and with reduced personnel costs.

[0062] According to a further embodiment, the recirculation conveyor line 5 can also be triggered to empty. The trigger can be triggered when the recirculation rate increases, for example, when the recirculation rate exceeds a threshold. However, this may increase the susceptibility to errors due to fluctuations in the error rate of the singulation system 6.

[0063] Further embodiments include the use of various recognition-based methods: According to further embodiments, the data of all recirculated objects is stored, for example: Time of recognition; Image information of the piece goods, a part of the piece goods or a piece goods cluster; Extracted data of the piece goods, a part of the piece goods or a piece goods cluster, such as one or more heights, one or more volumes, one or more dimensions, one or more colors, one or more types, one or more image fingerprints, one or more surface textures; One or more barcodes attached to the piece goods or to the piece goods cluster; One or more plain texts attached to the piece goods or to the piece goods cluster and extracted by means of text recognition; One or more conveyor speeds.

[0064] According to further embodiments, newly detected exception objects are compared with a list of previously recirculated piece goods within the relevant time window (e.g., 1 min + / - 20 s). By comparing all data, a weighted probability for the occurrence of recirculation is calculated. The position and orientation on the conveyor line, which may be configured as a belt, may change slightly upon repeated detection, as may the contour and surface structure of deformable piece goods complexes.

[0065] According to further embodiments, the collected data is reused, possibly modified, in algorithms of other sorting methods such as ArtID / Letter Fingerprint. Alternatively, a 3-strike rule or an n-strike rule can be applied, which stipulates that general cargo complexes must undergo at least two or more recirculations to be flagged as repeatedly faulty or repeatedly recirculated, respectively.

[0066] Decisions for statistical evaluation can be recorded.

[0067] Further examples include: Application of automatic recognition methods for object recognition; inclusion of all known information from previous decisions; maintenance of a database of recognition results from the last few minutes; algorithms for the weighted evaluation of various data and classification results; methods for reducing subsequent errors from automatic errors, particularly duplicate detection.

[0068] According to the examples, the following advantages can arise: Reduction in manual intervention; Statistical analysis and storage enables the identification of critical shipments / objects, which can be used, for example, for process optimization in packaging or for retraining the recognition system; Reduction in material stress that would occur during repeated recirculation; For example, shipments or other piece goods are protected; without automatic classification as having multiple errors, they would often be conveyed in circles on the recirculation conveyor line and thus be subjected to heavy mechanical stress. Increased throughput or maintained throughput; The method can also be applied to other systems with automatic recirculation (e.g. barcode noread, baggage, letter...); Increased process stability.

Claims

1. Logistics system (1) for piece goods, comprising a main conveyor section (2), a detection system (3), an ejection system (4), and a recirculation conveyor section (5); - wherein the main conveyor section (2) is designed to convey piece goods to and from the detection system (3) further downstream; - wherein the ejection system (4) is designed to feed piece goods, which have been fed to the detection system (3), from the main conveyor section (2) to the recirculation conveyor section (5); - wherein the recirculation conveyor section (5) is designed to feed piece goods, which have been fed to it by the ejection system (4), to the detection system (3) again; - wherein the detection system (3) is set up to generate a data record (33, 34, 35, 36) for a piece good complex (23, 24) fed to it, by means of which the piece good complex (23, 24) can be identified; - wherein the detection system (3) is also set up to detect an error affecting a piece good complex (24) fed to it and to cause the ejection system (4) to feed the piece good complex (24) affected by the error to the recirculation conveyor section (5) characterized in that - the detection system (3) is also set up to store at least those data records (33, 34, 35, 36) by means of which piece good complexes (23, 24) fed to the recirculation conveyor section (5) can be identified- wherein the detection system (3) is also set up to compare data records (33, 34, 35, 36) with one another and, in the event of a match between two or more data records (33, 34, 35, 36), to classify this match as evidence of a repeatedly faulty piece good complex (24).

2. Logistics system (1) according to claim 1, comprising a singulator (6) which is designed to singulate piece goods fed to the detection system (3) and the error, on the basis of which the detection system (3) causes the ejection system (4) to feed the piece good complex (24) affected by the error to the recirculation conveyor section (5), is an error in the singulation of the piece good complex.

3. Logistics system (1) according to any one of the preceding claims, wherein a data record comprises a selection of the following data: - Time of detection; - Image information of the piece goods, a part of the piece good or a piece good cluster, wherein the image information can also include several images from different perspectives; - extracted data of the piece good, a part of the piece good or a piece good cluster, such as one or more heights, one or more volumes, one or more dimensions, one or more colors, one or more types, weight, recognizable sub-elements (e.g. stickers, case rolls, closures...), position in relation to other piece goods, one or more image fingerprints, one or more surface structures; - one or more barcodes which are attached to the piece good or piece good cluster; - one or more plain texts that are attached to the piece good or piece good cluster and that were extracted using text recognition; - one or more conveyor speeds.

4. Logistics system (1) according to any one of the preceding claims, wherein the detection system (3) is set up to perform a preferably time-based search space restriction when comparing data records.

5. Logistics system (1) according to any one of the preceding claims, wherein the detection system (3) is also set up to determine a probability for the match of two or more threshold values and, if a threshold value is exceeded by the probability, to classify the match as evidence of a repeatedly faulty piece good complex.

6. Logistics system (1) according to any one of the preceding claims, wherein the logistics system is a sorting system and the data contained in a data record is additionally used for sorting the piece good assigned to the data record.

7. Logistics system (1) according to any one of the preceding claims, comprising a manual processing station (7), wherein the detection system (3) is set up to cause the ejection system (4) to feed a piece good complex affected by multiple errors to the manual processing station (7).

8. Method for operating a logistics system for piece goods, comprising the method steps: - conveying piece goods in the logistics system along a main conveyor section (2) to an automatic detection system (3); - automatic detection by the detection system (3) of an error affecting a piece good complex (23, 24); - Creation of a first data record (33, 34, 35, 36), by means of which the piece good complex (23, 24) can be identified; - due to the detected error, ejection of the piece good complex (23, 24) from the main conveyor section (2) to a recirculation conveyor section (5), which causes the piece good complex (23, 24) to be fed to the detection system (3) again; - wherein when the piece good complex (23, 24) is fed again, the detection system (3) detects the error again and creates a second data record (33, 34, 35, 36) by means of which the piece good complex (23, 24) can be identified; - detecting a match of the first and second data records (33, 34, 35, 36)and based on this classifying the piece good cpmplex (33, 34, 35, 36) as repeatedly affected by errors.

9. Method according to claim 8, wherein the piece goods fed to the detection system (3) are conveyed through a singulator (6) and the detection system (3) measures the quality of the singulation of the piece good complex.

10. Method according to any one of claims 8 or 9, wherein a data record comprises a selection of the following data: - Time of detection; - Image information of the piece good, a part of the piece good or a piece good cluster; - extracted data of the piece good, a part of the piece good or a piece good cluster, such as one or more heights, one or more volumes, one or more dimensions, one or more colors, one or more types, one or more image fingerprints, one or more surface structures; - one or more barcodes which are attached to the piece good or piece good cluster; - one or more plain texts that are attached to the piece good or piece good cluster and that were extracted using text recognition; - one or more conveyor speeds.

11. Method according to any one of claims 8 to 10, wherein the method step of detecting a match of the first and second data records comprises comparing the second data record with a plurality of stored data records (33, 34, 35, 36).

12. Method according to claim 11, wherein the detection system (3) is arranged to perform a preferably time-based search space restriction when comparing data records in order to select the plurality of data records (33, 34, 35, 36) with which the second data record is compared.

13. Method according to any one of claims 8 to 12, wherein the method step of detecting a match of the first and the second data record comprises automatically determining a probability for the match and automatically interpreting an exceeding of the threshold value as evidence for a repeatedly faulty piece good complex.

14. Method according to any one of claims 8 to 13, wherein the data contained in a data record is additionally used for sorting the piece goods assigned to the data record.

15. Method according to any one of claims 8 to 14, wherein the piece good complex classified as being affected by multiple errors is automatically fed to a manual processing station (7).