The concept of hindsight being 20/20 is no longer a viable excuse. Digital twins are the ultimate insight and when used properly, the benefits are substantial and unarguable. When you embrace digital twins from the design phase through the product’s entire lifecycle, everything goes up. This includes the quality of the data, the quality of your product and of course, a quality ROI. Flags become proactively raised, reducing design changes and avoiding potential failures. Unnecessary engineering costs, support costs, assembly costs and production costs are ultimately eliminated. Instead, they are replaced with better business decisions, better brand recognition and higher customer loyalty.
The Value of Information
Regarding design, production, manufacturing and troubleshooting of any product, literally anything will be possible with the use of digital twins. This is what we know. Think of digital twins as enormous mines and data is the new gold. What’s important now is spending the wealth wisely, without it going to our heads. We are stepping into a world where we are learning how to both manage and profit from digital threads, and there is simply so much of it…where do we begin?
Quality Produces Quality
A good place to start is on the production floor. The role of the smart factory is to maximize the benefits of automation. The end-product that is being sent out into the world is produced by assets that need to be equally reliable. These assets have your brand in the palms of their robotic hands and need to be as thoroughly monitored as the product itself. As everyone knows: garbage in, garbage out. The quality of the factory needs to match the quality of your product.
With digital twins, this is par for the course. Plant managers can track key performance indicators (KPIs) and trends being produced by their assets thereby improving quality and productivity overall. With the right software, Big Data is transformed into easy-to-read dashboards, giving users deep insight into their production line while facilitating machine learning. This includes yield trends, cycle time and throughput trends as well as measurement trends. By integrating this information into a company’s supply chain, any deviation from the norm becomes obvious, allowing expensive and time-consuming issues to be addressed efficiently or prevented altogether.
Example Issue 1: Product Defect and/or Recall
A product gets sent back for return or replacement due to performance failure. Today, often the process is to go back to the serial number, do some digging, ask around, review the product revision, etc. It becomes the world’s longest game of connect the dots, taking up valuable time and energy from people who would rather be doing anything else. By accessing the test data of the asset responsible for the product, hours of answers are provided instantly, including the magnitude of the problem.
Example Issue 2: Production Downtime
By analyzing the results and trends of the assets themselves, they are sending a message on how they are performing. Assume that when testing a product, it generally returns a steady average. Over time as production continues the averages of that product can shift but remain steady. Activity like this is likely trying to tell the user that something is off, and attention should be paid to the source. By having this kind of data provided in real time, scheduling proactive system maintenance becomes obvious, thwarting a shutdown catastrophe. Equally important, it can be done when it’s convenient.
Gold Standard of Manufacturing
There are infinite results that can be driven through the proper use of the massive amounts of data that can be collected. The key is filtering and managing the information, making it understandable and actionable. Integrating key data sources (ie. ERP, MRO, etc) together and leveraging the proper data management tools creates a compass allowing you to navigate through this pre-existing mine. All that will be left to do is collect the gold.