Fraunhofer Institute for Industrial Mathematics ITWMAt Fraunhofer ITWM, we are developing »RAPIDZ« a tool that uses Digital Twins directly in production. With our virtual models of physical systems, we monitor and control material flows, energy consumption, and machine performance in real time. In this way, we help companies to make their production lines more efficient, respond flexibly to changes, and avoid downtime through targeted maintenance.
RAPIDZ – Mitarbeiterin analysiert digitale Produktionsdaten am Computer mit einem Digitalen Zwilling
Digital Twins as the Centerpiece of Smart Manufacturing
At Fraunhofer ITWM, we are actively shaping production in the era of smart manufacturing. Comprehensive digitalization and the integration of intelligent technologies enable processes to be controlled more efficiently. Digital Twins play a central role in this: they record and analyze energy consumption, emissions, and operating data, thereby creating direct added value for companies.
© iStockphoto
Digital Twins for Efficient and Sustainable Production
Our Modeling Platform »RAPIDZ« – Resource Analysis and Process Integration Through Digital Twins
Digital Twins offer enormous opportunities, but their introduction is often complex and requires many unknown parameters, such as material parameters. With »RAPIDZ«, we make it easier for companies to use digital twins in production and overcome these hurdles with a modular design. Our digital twins mathematically map physical production systems and are continuously fed with real-time data from sensors and IoT-devices. This allows us to create transparency regarding plant conditions, productivity, energy consumption, and emissions, thereby increasing Overall Equipment Effectiveness (OEE).
Technical Setup of Production Lines within the Digital Twin
Even though each production line is set up individually, they all share a basic structure. We see them as a chain of machines that are clearly defined and connected to each other, sometimes linearly, sometimes sequentially in a branched manner. At the beginning, there are raw materials or preliminary products that are picked up by the first machine and processed step by step.
This generic structure forms the basis on which we develop our Digital Twins with »RAPIDZ« and virtually replicate the processes.
Every machine, every transition, and every material flow is recorded as a module and integrated into the Digital Twin. This creates a flexible model that can be adapted to different production environments. Our approach goes beyond pure machine logic: we integrate physical properties, operational parameters, and data streams.
Overall Equipment Effectiveness and Resource Optimization
This allows us to simulate, analyze, and improve processes before implementing changes in real production. Our Digital Twins can be used in numerous areas of focus. What does that mean in practice?
- Real-time monitoring for trouble-free operations: We continuously collect data from machines and production lines. This enables us to identify bottlenecks and malfunctions at an early stage and ensure the availability of the systems.
- Make predictions for better production decisions: By simulating various scenarios, we show how changes affect throughput and resource utilization. This enables companies to make more informed decisions and plan processes more effectively.
- Optimize production performance: We use data-based algorithms to determine optimal production speeds, adjust processes, and plan maintenance to increase overall equipment effectiveness.
- Continuously ensuring product quality: We continuously monitor product quality. We identify deviations early on so that companies can quickly take measures to improve products.
- Improving sustainability and energy management: »RAPIDZ« takes into account current operating data, electricity prices, weather forecasts, and contractual requirements. In this way, we help to optimize energy consumption and emissions, make efficient use of local generation and storage, and make production sustainable.
Practical Examples of Digital Twins in Action – Bottling Beverages or Assembly Line Work
In »RAPIDZ«, we describe each machine with an operating point-dependent »recipe«. We know how much pre-product goes into an intermediate product, how long processing takes, and how much time is required for start-up, shutdown, or switching between operating points. For example, a beverage filler needs a bottle from the conveyor belt, one liter of beverage, a screw cap, and a certain amount of time to complete a single product.
Simulation einer Getränkeabfüllanlage mit »RAPIDZ«: Die SCADA-Oberfläche zeigt, wie sich Produktionsprozesse in Echtzeit überwachen und steuern lassen.
Transport systems such as conveyor belts and forklifts can also be modeled in »RAPIDZ«. They do not process the product directly, but determine transport times depending on speed or cycle times. This allows us to integrate not only production itself, but also key logistics processes into the digital twin.
In addition to production machines, we also record auxiliary systems that are not linearly arranged in production but encompass several machines. This can be the case physically – for example, in clean rooms – or purely logically, as with a shared power or compressed air supply. We implement auxiliary systems in a standardized structure. Any number of machines can be connected to each component to exchange data and status information. For example, an air conditioning system collects the waste heat and humidity from all enclosed machines and, in return, feeds back the resulting temperature and humidity to them, depending on its own performance
Model Predictive Maintenance for Efficient Production Lines
In production lines, machines are often directly dependent on each other. If one component fails, it immediately affects all subsequent steps. With »RAPIDZ«, we can map such dependencies in the digital twin and run through scenarios: Do we continue until there is a backlog, or do we adjust the operating mode at an early stage? This allows us to develop strategies that are more energy-efficient or easier on the equipment. Using Model Predictive Control (MPC), we simulate controller decisions directly on the digital twin, evaluate their effects, and pass on the optimal setting as an assistance recommendation or directly to the machines.
Konzept für die verteilte Regelung mittels unserer Modellierungsplattform »RAPIDZ«
Optimized Operating Modes Using a Modular Structure
Thanks to the modular design of »RAPIDZ«, controllers can be distributed not only centrally but also across individual machines – while still achieving the overall plant objectives. We can take various optimization goals into account, such as scrap rates, product quality, or resource costs. This allows us to assign specific costs and quality indicators to each production step and find the operating mode that makes the entire line the most efficient, sustainable, and economical.
Smart Control of Energy Consumption and Plant Operation
In »RAPIDZ« we also map energy supply in a modular way. For each operating mode, we record the average consumption, which is usually known or measurable. Instead of simply stopping systems during peak loads or avoiding high-load times altogether, we incorporate additional information such as weather forecasts, electricity prices, or contractual limits. This makes energy management and sustainability integral parts of production operations.
Using the same principle, we integrate emissions such as waste heat, particulate matter, or CO₂ into the digital twin. This provides companies with a precise basis for aligning their production with legal requirements and making it more sustainable. »RAPIDZ« thus combines the optimization of energy and resource use with the reduction of emissions.
Project Funding
»RAPIDZ« is part of the »Re(Pro)³ project – Resource-optimized production through inline process and product monitoring«, which is funded by the Rhineland-Palatinate Ministry of Science and Health (MWG).
