Fast and dependable classification and assignment of pallet types at Rossmann

The more efficient its work processes, the more successful a company can be in the long run. As one of Europe’s largest health and beauty chains, Rossmann learned this valuable lesson years ago and is optimizing many of its own processes for this very reason – including those used in receiving. Rossmann works with a wide range of suppliers, which means the company handles a broad variety of pallets on which its products are loaded. Some of these pallets are part of a pallet deposit system for which they first have to be classified and correctly assigned. This manual process is both time consuming and labor intensive. It is also a job whose high potential for errors cannot be ignored.

The products from Rossmann are delivered on different pallets, which must be classified.

Rossmann now tackles these jobs with the help of two SICK solutions: the Pallet Classification System (PACS) and the Tag-LOC System. Thanks to these systems, Rossmann was able to automate the classification of pallets in receiving. They were also able to automate their assignment by using a real-time localization system in the company’s material handling equipment. This change not only lowers the company’s costs, but also optimizes process quality, pallet-inventory transparency and use of resources.

Rossmann asks the deposit question

With its 62,100 employees and 4,741 stores, Rossmann has grown to become one of Europe’s largest health and beauty chains since its founding in 1972. Its assortment comprises 5,000 store-brand products and 18,000 products made by other companies. Operations in receiving are fast-paced as a result. In the past, delivered pallets were manually received, classified in accordance to their deposit significance and assigned to each respective supplier – a time-consuming, error-prone process. In searching for ways to streamline the entire system, Rossmann turned to SICK.

SICK finds the answer

SICK’s solution combines two systems: the pallet-classification system PACS and the real-time localization system Tag-Loc. The integration work was performed by BSS Bohnenberg, a company with which SICK has successfully carried out other projects in the past as equal partners.

Rossmann uses the Pallet Classification System, which automatically classifies incoming pallets.

Classification made easy

The PACS quality control system, which can be installed on many places of conveyor units thanks to its low space needs, handles the job of automatically classifying pallets upon their arrival in receiving. While the pallets are traveling along the conveyor system, two 2D machine-vision cameras, the midiCam2, take photos of the outer pallet blocks and create RGB images of them. An optional third color camera can be mounted above the conveyor system to document the load on the pallet.

The system evaluates these images with the help of an artificial neuronal network to identify the type of pallet, e.g., EPAL, CHEP or UIC. The system can be taught with other types at a later stage. Potential deviations in the quality of the branding are considered during the network-training process, something that is reflected not only in the success rate but also in the robustness of the evaluation.

The collected data is then transmitted to a higher-level warehouse management system to determine which deposit-relevant and non-deposit-relevant pallets have been delivered.

Precise assignment thanks to real-time localization

The Tag-LOC localization system is based on ultra-wideband sensor technology and enables high accuracy.

Tag-LOC is a system used to determine positions. It uses ultra-wideband (UWB) sensor technology to facilitate real-time localization of material and vehicles with decimeter-range accuracy. For Rossmann, BSS installed 10 antennas (or anchors) in the receiving area and 10 tags on electric pallet trucks whose batteries keep the tags supplied with power. Once a supplier has been assigned to one of the 10 electric pallet trucks, the appropriate tag sends movement signals via UWB to the antennas. The antennas then transmit the signals to the software platform Asset Analytics that has been installed on the customer’s server. The pallet truck is positively identified there, and its exact position is determined.

Virtual counterparts of the four conveyor lines were defined in the software at an earlier stage. When a pallet truck is located in one of these four areas, both its ID and the ID of the conveyor line in the warehouse management system will be transmitted to the higher-level system. In this way, the pallets classified by PACS can be assigned directly to the respective pallet truck – and via its ID indirectly to the supplier.

Localization systems

Exact localization, tracking and management of assets

Tag-LOC System

Quality control systems

Quick and reliable classification of different pallet types

Pallet Classification System

Efficient interplay

The combined solution pays off for Rossmann in many ways. The system can be cost-efficiently maintained and serviced thanks to its simple integration into the existing warehouse environment and the standard sensor technology of PACS. Because the job is no longer done manually, employees gain time that they can use to concentrate on processes that generate more value. Unlike other conventional image-processing solutions, the use of deep-learning technology in the SICK solution requires no detailed programming knowledge, because the system was trained with specific examples. As a result, SICK was able to develop the pallet-identification system for Rossmann comparatively easily.

At the same time, Tag-LOC prevents erroneous assignments thanks to its link to Asset Analytics. Should several pallet trucks be identified on one conveyor line at once, Asset Analytics will send an alarm directly to the higher-level system. This step ensures that a single conveyor line will be loaded with the products of only one supplier.

Trailblazer for future projects

Since successfully automating the entire product-receiving process during ongoing operations, Rossmann has profited from an efficient, fast and reliable classification and assignment system for different pallet types. For the chain of health and beauty stores, this means cost savings produced by correct deposit billings, optimal work processes and total transparency of the entire pallet inventory, including clear assignment to suppliers. Martin Gräb, Managing Director for Marketing and Sales at the integration partner BSS, expressed his satisfaction with the system: “With the automated product-receiving system, BSS and SICK have established an individual and highly innovative solution in the marketplace that will be a trailblazer for future projects.” Rossmann has already implemented one of them and expanded the Tag-LOC System by integrating three forklifts.