6 minute read 24 Feb. 2020
Automotive production line welding car bodies

How digital twins give manufacturers a real-world advantage

Authors
Sven Dharmani

EY Global Advanced Manufacturing & Mobility Supply Chain Leader

Passionate about transforming supply chains. Problem solver. Curious and collaborative. Avid traveler, scuba diver and car enthusiast.

Sachin Lulla

EY Americas Advanced Manufacturing & Mobility Consulting Sector Leader

Internet of Things strategist. Digital influencer. Sector-focused thinker. Keynote speaker. Proud husband and father.

6 minute read 24 Feb. 2020

These virtual replicas of a physical product, equipment, process or the supply chain help you monitor, simulate and optimize performance.

If you’re creating a new tire, you’ll probably mold a prototype and test it out on the road, then update it with refreshed versions and more safety checks.

If you want to boost production capacity, you can rely on your instincts to try to predict the impact on your assembly lines and react to bottlenecks as necessary.

And if you’re worried about a disruption to your supply chain from a natural disaster, you can consider weather forecasts and other relevant data to create contingencies — and then hope for the best.

But in the Transformative Age, manufacturing companies — particularly those in the automotive sector — have another option that is more nimble, more cost-effective, more cutting-edge and more grounded in the real-time reality of how your business actually operates. It’s called digital twin — a virtual replica of any physical product, equipment or asset, or the supply chain, that you can use as testing grounds for monitoring, simulating and optimizing production, quality and operational performance.

A digital twin could be a virtual version of a tire or a passenger vehicle, the methods for how these items are built or the entire production line. It could even be an entire factory, a network of plants or the end-to-end supply chain. Together with disruptive technologies, such as the Internet of Things (IoT), artificial intelligence, machine learning, process and data mining, and even augmented reality, digital twins are stretching the bounds of what’s possible, enabling the fourth industrial revolution to transform manufacturing for the 21st century.

Let’s explore how they work, and how they can be applied in your operations.

How they work

When looking at factories and the manufacturing process, companies create a digital representation of a piece of equipment, then IoT sensors collect real-time data on its performance and transmit it to servers, either on the premises or in the cloud. With that data, you have an exact digital replica of that equipment and its role in your operations. The more data you get from more sources, the more insights you can glean into how your factory runs, through data analytics and machine learning, and you’re equipped to run simulations. And you can go a step further in your visualizations through virtual reality and augmented reality.

You also have the opportunity to bring in more data from other sources, in other contexts, like logistics information in your supply chain, and serial numbers from components in transit. With the right data, your digital twin becomes a laboratory for exploration, where hypotheses can be tested and forecasts can be sharpened.

A host of related opportunities exist for your products as well, whether you’re still developing them or wondering how they’re being used in the market and can be improved. By visually representing one asset or group of assets in the digital realm and enriching it with data based on the physical world, you’re able to tap into a wide variety of transformative possibilities. Here are some of them.

Putting digital twins to work

Product testing: Consider the case of tires from the start of this article. Through a digital twin, you no longer need to wait for performance data from vehicle trials to determine its quality and performance.

You can start to simulate performance of your product — such as a new tire in different weather conditions — to proactively improve design and virtually experiment with different compounds and raw materials to optimize the performance in different conditions.

You can predict the performance of the product and make changes in real time, helping you determine the optimal configuration (for instance, to meet tire safety standards). You can even model the replacement demand from a multitude of performance factors, such as driving behaviors.

Adding manufacturing capacity: Every time you introduce a new product, you disrupt the existing assembly schedule. Before you spend money to commission new equipment and add production capacity, you can simulate the impact on your production capacity and schedule based on the unique characteristics and demands of what you’re manufacturing.

For example, perhaps an automaker can make only a certain number of convertibles in a day because of the added burden of installing the wiring for the roof. While you can add people to do those additional tasks, you can’t overcrowd the assembly line. That has to be balanced with how many other cars are you making — say, those with sunroofs — and what percentage of each you can make in a day, based on order patterns.

Through a digital twin, you can simulate your assembly line based on the options, even before you’ve planned your week and see where the process issues lurk. You can even do real-time testing of virtual equipment from an engineering perspective before it’s installed and hard-wired.

Plant infrastructure: How about monitoring a plant without physically being there? With a digital twin, you can do that and more, like walk the shop floor of a plant in Mexico, through virtual reality, and train workers remotely from a laptop. With a digital twin the equipment doesn’t even have to be physically installed to show others how to use it.

Asset maintenance: By monitoring real-time data from IoT sensors in the manufacturing process, your digital twin can also model when maintenance will be needed for critical equipment, or it can even be used to determine the health of an entire production line, factory or network of factories. In a 2018 report, Gartner estimated that businesses (and consumers) will save US$1 trillion each year in asset maintenance by using IoT through digital twins.

Targeted recalls: By tracking key components that go into your product and their serial numbers in a digital twin — providing full traceability — you don’t have to recall tens of thousands of products in the event of a problem. You’re able to narrow your efforts only to those directly impacted — for instance, by pinpointing a shipment of components that experienced corrosion in transit from overseas.

Retail: Digital twin even helps in consumer-facing realms. For instance, a digital twin of the vehicles you produce, with 3D visualization, allows a potential buyer to change the color of a car to one that he or she prefers when shopping online. The same goes for different vehicle options for the interior, providing a richer and more personalized customer experience.

Aftermarket/consumer service: In the used market, consumers have to guess the health of a vehicle, or perhaps look at a Carfax. But if you give the vehicle a digital identity (with a digital twin and blockchain), you can then start to track how often it was serviced, answering questions like: When parts were changed, did they use original equipment manufacturer (OEM) parts or aftermarket parts? Was there an accident that did not get registered, based on an impact analysis?

The twin holds all real-time performance, sensor and inspection data, as well as service history, configuration changes, parts replacement and warranty data.

Supply chain: Digital twin allows you to monitor and simulate events that impact your supply chain — for instance, a natural disaster or a trade dispute with a key country — and how you can proactively react to them.

For instance, if you want to simulate the impact of reduced raw materials from China, a digital twin can help you decide how to fulfill customer commitments efficiently. You can start to model various data points — such as how long the component spends in transit, or whether it’s perishable — and start being more proactive instead of reactive.

But this only scratches the surface of what you can do with a digital twin. In a future article, we’ll explore the supply chain issue more deeply and how other technologies can revolutionize it from end to end.

Summary

By visually representing a physical product, equipment, process or the supply chain in the digital realm, and enriching it with real-time sensor data based on the physical world, digital twins are stretching the bounds of what’s possible, offering a virtual testing ground for smarter products, manufacturing and optimized supply chains.

About this article

Authors
Sven Dharmani

EY Global Advanced Manufacturing & Mobility Supply Chain Leader

Passionate about transforming supply chains. Problem solver. Curious and collaborative. Avid traveler, scuba diver and car enthusiast.

Sachin Lulla

EY Americas Advanced Manufacturing & Mobility Consulting Sector Leader

Internet of Things strategist. Digital influencer. Sector-focused thinker. Keynote speaker. Proud husband and father.