June, 2017 Machine Learning... Continued from page 56
aging and storage location. The item is measured using a
light grid that is passed over the item in one movement and provides accura- cy to the nearest millimeter. In con- trast to lasers, light grids can process partially-transparent, black and reflective materials and operate at a higher speed. The system delivers cor- rect and consistent data that can be transferred by an interface directly to the ERP system and processed. Reliable information about the
volume, size and weight of items makes it possible to provide correct master data for analysis, consistency checks and data synchronization. A volume light grid forms a simple link between the sensor system and the database. It is a suitable smart factory solution for small- and medium-sized companies with a clear application landscape and limited numbers of products, and can be quickly and easi- ly integrated into existing systems.
Integrated System for High Volumes
Where large volumes of goods
are shipped in individual packages, incorrect information about the size and dimensions of the products can
The data is stored in the ERP
system as soon as the package has passed. It can also be linked with the selected delivery quota. By matching the data and the current quota capac- ities, the control system can identify that shipping company “X” has free capacity and can attach the correct shipping label to the next package that falls into a specific category. The individual package data is
constantly synchronized with the ERP, logistics and merchandise infor- mation systems in real time. The process of reducing the shipping costs and choosing the best possible ship- ping company can be fully automated. A logistics dashboard can be
created based on this complete data set that can answer several analyti- cal questions to support managing the logistics process. In addition, intelligent evaluation software can be used to strategically plan manage- ment activities and to negotiate bet- ter conditions for delivery quotas with shipping service providers. The result is a complete smart
factory workflow at the conveyor involving large volumes of packages. Sensors and hardware are integrated into the overall system, reducing costs.
Fully-Automated Machine Learning
One of the major challenges fac-
Through networked dashboards, machines are monitored and controlled.
lead to additional costs. Using unnec- essary extra packaging materials and shipping packages containing empty space is expensive. However, even more money is wasted if the wrong shipping company is chosen. To find the best possible price-
to-performance ratio among the many shipping service providers and their various offerings, companies that dispatch large volumes of goods often negotiate specific service pack- ages or quotas in advance. When a package passes along
the conveyor belt, a decision must be made in real time about the best quota for that package and how it can be shipped most cost-effectively. Based on the volume, the weight and the resulting decision, the package is automatically labeled for the appro- priate shipping company and quota. If the data that has been record-
ed is not correct, the package cannot be matched with the best shipping solution. As a result, the quotas that have been negotiated will not be used effectively and the shipping process will become inefficient. With a large number of items,
minor problems can quickly add up to large sums of money being wasted. This can be avoided if the data is being stored centrally and captured correctly. An advanced logistics 4.0 solu-
tion that allows packages and quotas to be fully-automated consists of vol- ume light grids integrated into the conveyor and the machine control system. Scales record the weight of a package and the volume light grid captures its dimensions, all at a con- veyor speed of up to 2 m/s (6.6 ft/s).
ing large e-commerce retailers and shipping and courier companies is sorting packages to ensure that they are subsequently processed correctly. Machine sorting is almost impossible because of the wide variety of differ- ent types of packaging with different shapes and sizes. Other problems include the
extensive range of packaging materi- als and the fact that the packages are often stacked or heaped on the con- veyor belt. Large volumes of goods still need to be sorted manually. This leads to considerable added costs and slow sorting procedures with high error rates. The solution is machine learn-
ing. First, a large amount of image data is supplied to an algorithm, for example, tens of thousands of images of packages stacked on a conveyor. Initially, this image data is classified manually. The algorithm is told, “This is a package.” It is then able to sort the packages into categories based on their individual features. The classification process using
a neural network is a multistage process requiring a large amount of computing power, separating the data during several phases to classi- fy it. The machine learns the param- eters that identify an object as a package, enabling the algorithm to classify individual objects independ- ently in the future. Machine learning is used where
conventional imaging has reached its limits, such as sorting a variety of package types. Three-dimensional scanners are not an adequate means of identifying the shape of the objects and operate only at low speeds. For a system with a robotic picker, this leads to inaccurate sorting, incorrect routing of packages, system failures, waiting times, and time-consuming manual follow-up work. By contrast, systems based on
machine learning can accurately iden- tify the parcels and their shape using the classification scheme that they have learned, and the robot can carry Continued on page 61
www.us-tech.com
Page 59
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88
