Beverage bottling

Thomas Becker, Stephan Birle, Georg Götz, Tobias Voigt | Günter Herkommer,

RoboFill 4.0 - breaking up rigid system concepts

Current (filling) concepts in the beverage industry are generally based on rigid automation structures. The RoboFill 4.0 project aims to change this - with robotics playing a central role.

© Fotolia, industrial view

The provision of personalized and customized products is an unavoidable trend that is also increasingly shaping the food production market. Individual desires can be characterized by a wide range of motivations, such as changing health awareness or the pursuit of naturalness. Highly successful marketing strategies for products such as MyMüsli clearly demonstrate the existing and growing market potential. MyMüsli already offers its customers the opportunity to put together their own muesli via an Internet-based store system and then order a customized composition.

The focus of such concepts is always on the direct integration of customer interests in the production and picking process of a food product. This means that the consumer contacts the producer prior to production and the production process is carried out according to the wishes expressed. Delivery is also made directly to the end customer, so that additional trading stages or storage processes can be largely dispensed with.

When it comes to putting such concepts into practice, however, the production and process control structures used to date quickly reach their limits. This is because today's production structures are not designed to ensure the required quantity and format flexibility economically and sustainably. This is particularly evident in the beverage industry: the filling and packaging systems used here today are complex or rigidly interlinked lines that consist of several machines and modules and are designed for high repetition rates of individual products. The industrial filling of beverages in particular is currently carried out almost exclusively in carousel machines with height/volume dosing systems or weighing systems. With a correspondingly high level of specialization, such systems can achieve an output of up to 80,000 bottles per hour. However, quick or even individual changes of the formats or products to be processed are not possible, taking economic aspects into account.

The existing automation concepts and current production processes are correspondingly inflexible. Although the production systems are networked with each other in the sense of the IT pyramid that has been established for decades, their work is rigid in sequence and linked to pre-defined structures. This also applies to the individual production units. They cannot be operated autonomously, nor can they be networked with cloud-based services without major effort in order to be able to react individually to corresponding consumer, process or logistics requirements.

In short: due to the rigid mechanical and IT interlinking of the machines, there are strict limits to individualization, even in the sense of modularized production lines. In this respect, there is currently no adapted, dynamic filling concept for small and medium-sized companies that would allow format and batch size flexible production with just a few bottles or even one-offs.

Advertisement

Figure 1: The flow diagram of a rigidly interlinked returnable beverage filling line shows the current state of the art in beverage filling.

© Technical University of Munich

This is precisely where RoboFill 4.0 comes in. The aim of this research project is to develop a completely new, flexible automation concept for the industrial provision of customized beverages that can be expanded modularly to include additional production units. A schematic representation of the concept is illustrated in Figure 1: All system components are to be designed as cyber-physical system components. In line with the 'Internet of Things' concept, these are able to communicate with each other and with higher-level systems using network and cloud technologies.

In addition to the provision and development of an adaptable and flexibly expandable filling and material flow concept, the focus is on a decentralized control concept that coordinates the modules in the system. Customer requirements, production orders and the associated production planning and control are synchronized via a virtual image. Communication with the so-called digital shadow and the associated exchange of planning-relevant data, such as the degree of utilization of a filling module, enables situational adaptation of the production process to optimize it.

The product also has a virtual representative, which interacts with the individual modules so that it can navigate itself through the filling process. This also makes it possible to dynamically design the local and temporal representation of the individual system components. In contrast to the rigidly designed and controlled production lines of the bottling and beverage industry, the project aims to create highly flexible processing stations that are intelligently controlled by the product. To achieve this, the project partners are developing a technology platform based on a digital ticket. Using this, all components are designed to be self-sufficient and are able to communicate continuously with each other and exchange the current module status. The ticket, which is generated by the customer order, determines all process sequences as well as the material flow and any consumer interactions in a constant exchange of information via the cloud. As part of the research project, a demonstrator system is being created that includes all the necessary functionalities.

Agent-based control concept

Figure 2: Schematic representation of the filling concept for the individualized provision of beverages.

© Technical University of Munich

Up to this point, it can be said that The future-oriented Industry 4.0 project initiated by the Federal Republic of Germany offers ideal conditions for opening up new, customer-oriented and flexible value chains for the brewing and beverage industry. This involves merging plant technology components within a network with elements of IT and software components that communicate decentrally via the Internet using a data infrastructure.

A central aspect of this new production philosophy is the decentralized self-organization of production, which requires a product-based control process and thus heralds a fundamental change in the currently dominant manufacturing structures. Such an approach can be implemented with a so-called multi-agent or software agent system, in which each relevant production component represents an intelligent, communicating and purposefully acting object. The intelligent objects have goals that they must achieve independently. In addition, they have internal states about which they can provide information and exchange information with other intelligent objects or humans in production.

To implement the project, it is divided into several areas. The first step is to design the physical production modules of a filling concept (e.g. printer, rinser, filler, capper) as autonomous modules with CPPS capability. A particular technological challenge here is the creation of a flexible filling system. Direct inkjet printing of the bottles is planned for the customized design of the bottles (Till company). In a second core area, a flexible material flow system is used, which takes on the role of an internal logistics service provider. Special industrial robots with intelligent gripper systems realize the function-specific handling at the processing stations. A major task in this respect is the flexible design of the system and control technology in order to be able to react to all circumstances of a context-specific change in the production environment.

Figure 3: RoboFill 4.0 concept overview: The interaction between the customer portal, order system and production process, which results from the interaction between the physical and virtual production environment in the cloud, is shown.

© Technical University of Munich

As a result, the robots used perform a wide variety of handling tasks, which are called up and determined by the goods to be handled themselves. Above all, the gripper technology must be able to adapt to these different processes, as it represents the link between the product and the robot. The focus is on enabling the gripper to adapt quickly to the task, for example in terms of stroke and force. In addition, the handling system should be able to verify the parameters of the bottle, such as its shape, size and weight. The gripping system receives the information about the product to be gripped from the agent system and checks the data via the integrated sensors. In the overall context, this creates a flexible handling system in which the product is checked during the handling process.

The third core area involves the development of the production control concept and the synchronization of the individual material flow and system modules. In line with the guiding principle of production control 'through the product', a software agent system is designed and aligned to the application purpose of robot-assisted filling. In this respect, all physical modules of both the production and the material flow system are to be represented virtually with the help of software agents and the physical production environment is to be synchronized with the virtual image in terms of information technology. The virtual production environment is made up of cloud-based services, which include the agent system, a web-based customer portal and order and data management with a database system. The entire process, including the material flow sequence, can be mapped, coordinated and transferred to the real, physical system components using the cloud-based services.

Each module of the cyber-physical filling system (CPA) logs in with its specific properties and capabilities (skills) and its position data is recorded via the material flow system. The physical modules and the software agents initially only provide basic functionalities in relation to the overall system and are only supplied with production orders and production knowledge by linking the services offered, which correspond to their capabilities. The product-specific parameterization of the processing modules is carried out by the digital shadow of the bottle to be produced, which communicates the parameters directly. To ensure the traceability of all processes, the process data is also communicated and stored by the module.

Finally, an online customer portal is created for the direct integration of the consumer into the desired filling concept, which enables the customer to configure (e.g. different beverages, different containers, customer-specific equipment and design) and order individual beverage products.

Authors:
Prof. Dr. Thomas Becker is full professor at the Chair of Brewing and Beverage Technology at the Technical University of Munich;
Stephan Birle coordinates the RoboFill 4.0 project at the TU Munich;
Georg Götz is Group Leader Flexible Plant Technology at Fraunhofer IWU;
Tobias Voigt is a research assistant at the Chair of Food Packaging Technology at the Technical University of Munich.

The project partners

With the iAgent multi-agent system, Infoteam Software is contributing a core component for decentralized production control to the RoboFill research project. Due to the very close interweaving of hardware and software components, there is also interdisciplinary cooperation between various experts in brewing and beverage technology, filling and packaging technology, robotics, mechatronics, automation technology and information technology who have been proven in their respective fields of activity for many years. Specifically, these are the companies Krones, Beckhoff, Proleit, Zimmer, Siemens, Till and Yaskawa Europe as well as the Weihenstephan state brewery. The scientific and administrative coordination is carried out in cooperation with the Chair of Brewing and Beverage Technology, the Chair of Food Packaging Technology and the Fraunhofer IWU (RMV project group).

  • Xing Icon
  • LinkedIn Icon
Advertisement
Advertisement

You might also be interested in

Advertisement
Advertisement
Advertisement
Advertisement

Miba

The first steps towards digitization

Real-time transparency in the material flow: this was the goal set by Miba when it set out to digitalize its internal logistics processes. But how successful was the close link between ERP and MES in the end? - A field report.

read more...
Advertisement
Advertisement
Advertisement

Big Data

Online machine data under control

Turning huge amounts of data into valuable information - how can this smart industry approach be implemented? Linking PC-based controllers with Matlab and a cloud-based IoT analytics service can be a viable approach.

read more...
Subscribe to our newsletter
Advertisement
Back to home