Why Has BMW Created A Virtual Model Of A Real Factory?

The shrinking cost of sensors has fueled a boom for the Internet of Things (IoT). It is a subject we have often explored on this blog, in particular how it has enabled the increased use of digital twins.

Digital twins are used across many industries, but the most well-known example is aircraft engines. Every engine manufacturer has a virtual version of every individual engine. The virtual engine is constantly updated by sensors on the real engine and so the digital twin can help to predict problems and can be used to plan maintenance.

But digital twins can be used to model entire environments – such as a complete factory. This has the advantage of also allowing for experimentation and strategy to be tested virtually.

Process optimization can be tested without needing to change operations in the real factory and because the twin completely and accurately reflects the real factory, the process is reliable. Outcomes discovered in the virtual environment can be expected to take place in the real factory environment. This reduces the risk and cost associated with making changes to a production environment.

But there are a number of other important benefits to a company that builds a virtual version of a real factory.

• Cost Savings. Identify inefficiencies and optimize workflows before making costly real-world changes
• Improved Production Efficiency. AI-driven insights reduce bottlenecks and increase output with fewer resources
• Predictive Maintenance. Prevent breakdowns before they happen, minimizing downtime.
• Faster Innovation and Scaling. Test new products, layouts, and automation strategies in a risk-free environment
• Sustainability and Energy Optimization. Simulate and reduce waste, emissions, and energy consumption
• Enhanced Worker Safety and Training. Use simulations to train employees on best practices and safety protocols

The German car brand BMW has used digital twin technology to create a completely virtual simulation of their main car production factory. This allows the company to optimize how the factory works without impacting on production.

But there are new developments in artificial intelligence (AI) that are making the use of virtual factories even more important and interesting.

The virtual environment can become a test bed for AI and robotics, allowing for innovation to be tested before moving to the real factory environment. For example, a factory using autonomous mobile robots can test them virtually to ensure that the robots are using optimal routes around the factory.

This article from chip manufacturer Nvidia explains: “Developers can also use these physically accurate virtual factories to train and test AI agents capable of managing their robot fleets, to ensure AI-enabled robots can adapt to real-world unpredictability and to identify streamlined configurations for human-robot collaboration.”

While virtual factories are not yet common, they are an important emerging trend that utilize a number of existing capabilities and areas of technology such as the IoT and digital twins. It is likely that this trend of ‘virtual manufacturing’ will grow to the extent that most factories will eventually have a virtual replica.

For examples of IBA expertise in digital twins and the IoT, please click here. Follow IBA Group on LinkedIn for regular updates and comment.