A learning factory in Darmstadt, Germany, tries to find out how Industrial Internet of Things can work for medium-sized companies with old machines and processes.
How does the Industrial Internet of Things (IIoT) work for medium-sized companies?
The path to converting information into a digital format is not easy for small and medium-sized companies. For that reason, one of five IIoT competency centers for medium-sized businesses funded by the German Federal Ministry of Economics was set up at Darmstadt Technical University (TU). Here, companies can learn about the methods used to achieve digitization.
The "Center for Industrial Production" (CiP), a process learning factory, is set on the edge of the Lichtwiese Campus of Darmstadt TU. Inside is very similar to what is seen in actual manufacturing companies: The production environment is representative of many industrial businesses that work with machines. Within the learning factory of the Institute for Production Management, Technology and Machine Tools at the TU, people are allowed to test out the production system in place there. The goal is to recreate the development of an existing production system in the way of the IIoT in order to improve current production processes.
"We didn't just build a learning factory from scratch here, we are developing it at the level that can be found in many small and medium-sized companies. That's because modern production will not be created as an ideal starting from nothing. Instead existing devices and processes will be optimized from a previous state."
Andreas Wank, scientific employee at Darmstadt Technical University and project manager
"Compared to similar learning factories, Darmstadt TU has a feature that sets it apart," explains Andreas Wank, scientific employee of the Institute for Production Management, Technology and Machine Tools, who is responsible for the project. "We didn't just build a learning factory from scratch on a grassy meadow here, we are developing it at the level that can be found in many small and medium-sized companies." In other words, modern production will not be created in an ideal form starting from nothing. Instead, existing devices and processes will be optimized from a previous state. The European Union is among the sponsors for the "Efficient Factory 4.0" research project.
The use of digital technologies in production and work processes offer promising opportunities for increased competition and opening up new markets. The IIoT competency center in Darmstadt is also the central contact for small and mid-sized companies. It offers a complete range of practical training at no cost, from getting started with the Industrial Internet of Things to support for concrete solutions. The focus is on five areas: IT Security, Work in Times of IIoT, New Business Models, Energy Efficiency and Efficient Value-Added Chain Processes.
There is great potential for improvement in human/machine interfaces. Often, data entries still need to be made by hand in manufacturing processes and quality documentation is still on paper. Thus, the first step for most companies is automatic and digital data acquisition. This "digitization" forms the preliminary stage for implementing the vision of the IIoT.
The basis for improvement is always an analysis of the actual state. In the case of the process learning factory at Darmstadt TU, this means a machine park with a control unit dating back to 2005, with no interface compatibility and no dedicated system of sensors. An interface upgrade would likely cost about €20,000 per machine – a sizable investment.
But even without this upgrade, old machines can be partially networked. For example, an adapter was arranged on a lathe between a sensor for coolant fill level measuring and the machine control unit to provide bi-directional access via a tablet. The dial gauges of final product inspection for chipping production are networked with a central data acquisition system which detects whether the product meets requirements and sends automated error messages. Current transformers have been subsequently attached to other devices and they can be used to determine the load at which the machine is running. The data can be accessed through the central control system.
Tablet computers with a control interface, partially developed internally at the TU, are also used at various places in the learning factory as well as for remote access. This shows how human-machine interfaces can be optimized even in an existing production environment.
"The data from manufacturing is recorded and forwarded vertically in the company to the responsible employees."
explains Andreas Wank. If there is a problem with quality control, for example, the production manager receives a message on his tablet or smartphone and is able to stop the process. The responsible employee in manufacturing can click on the tablet to start a video conference and point out the problem.
This example already shows how competencies of employees will have to be shifted in the course of setting up the Industrial Internet of Things. As machines increasingly take over routine tasks and also control themselves, humans increasingly will need to intervene when there are problems. The role of employees will thus become decision makers and problem solvers.
For manufacturers that incorporate the Industrial Internet of Things, employees log in and receive the information they need. Even a batch size of one will no longer be a laborious special case, as demonstrated already by the TU learning factory. Different assembly videos can be accessed and projected on the white work tables for each employee depending on qualifications. The videos are recorded for highly variable manufacturing with real products and can be varied with a modular system. The goal of the IIoT is to simplify the process to the employee, not the other way around. That makes work easier and speeds up processes.
In addition to networking humans and machines, the IIoT also involves networking devices with each other. One keyword is "M2M communication" (machine to machine). First, products are fitted with a transponder, chip or code that is read on each machine and can be enriched with additional information. In the Darmstadt learning factory, an RFID system for wireless data transfer between machines and products is already in use – with some minor exceptions. For example, the raw material at the beginning does not have its own transponder yet because it would not withstand the milling. Instead, the small load carrier in which it moves through the hall can be tracked. This way, the path of the component through manufacturing can already be mapped, thereby creating the basis for an integrated quality control system. When combined with additional information that has been collected, this makes it possible to calculate the energy consumption per product. Then potential for savings will be easy to recognize.
A key element of this networked production is the control of the measurement chain in which the data of all sensors in use comes together and is processed. What that might look like can be explained using HBM’s PMX industrial amplifier as an example. The industrial amplifier is one of many devices from different manufacturers that bring digitized production on the university campus to life and is especially well suited to upgrading existing production chains.
The intelligent data acquisition system monitors and controls the entire production measurement chain (referred to as "condition monitoring"), ensuring that important supporting processes such as quality management and maintenance are optimized. PMX is also equipped with web-based software with a modern user interface.
Production systems can be networked in real time with PMX using an integrated Industrial Ethernet connection. PMX also uses electronic data sheets, or Transducer Electronic Data Sheets (TEDS) to detect sensors in the measurement chain and is able to parameterize them so that everything is ready to use again. Failures, deviations and redundancies are also detected, reported or bypassed. Processes and use of staff resources become more efficient. The system also has the ability to "learn" and adapt. The goal is self-optimization based on assigned key figures. Using modern systems in this way helps to lower manufacturing costs while boosting quality and speed.