I recently had the pleasure of sitting down with Maryah Merchant, Product Marketing Manager at Tecsys, Joe Vernon, Principle Business Consultant at EPAM Systems, and Ram Gopalakrishnan, CEO at Bricz, for a discussion on digital twin technology and the role it has to play in the future of our industry.
Use Cases of Digital Twin For Retail
In simple terms, a digital twin is a complete, virtual replica of a physical asset or process. It’s created by pairing comprehensive real-world IoT data provided by sensors, telematics and other connected devices with extensive machine learning analysis to create a perfect digital representation of its real-world equivalent.
As the retail industry undergoes an increasing amount of automation and transformation, digital twin technology can be exceptionally powerful. By creating a virtual model of their products and processes, businesses can identify potential issues before they occur, field-test product rollouts in a risk-free environment, and optimise future performance without the need for physical prototypes.
Having a digital twin of the business allows its leaders to introduce “what if” scenarios into their operations without creating disruptions or taking on the cost associated with assembling focus groups or piloting a change at a physical store. In this way, having a digital twin is like holding up a mirror to your business, making it possible to carry out detailed scenarios within a highly controlled environment, and get answers that are highly grounded in reality, without impacting the real-world business.
The Role of Digital Twin In Retail Supply Chains
Digital twin technology has the potential to rapidly transform the retail supply chain over the coming years, enabling businesses to optimise their logistical processes and streamline their operations. By creating digital replicas of their physical supply chains, businesses can gain valuable insights into their warehousing, distribution, and network optimisation strategies, which can lead to increased efficiency, reduced costs, and improved customer satisfaction.
One example of a company that is very actively working to push the limits of digital twin technology within retail is Lowes. Last September, the home improvement giant debuted a fully- digital replica of one of its stores, which associates could use to both visualise and interact with store data in an almost tangible way. This, in turn, opens the door to better optimising both retail operations and logistics through intelligent stocking of in-demand products, accurate forecasting of customer demand, and more proactive management of overstocks and out-of-stocks throughout the supply chain.
Overall, digital twin technology is a powerful tool for businesses in the retail sector, enabling them to optimise their logistical processes, reduce costs, and improve customer satisfaction. By creating digital replicas of their physical supply chains, businesses can gain valuable insights into their operations and make data-driven decisions that lead to greater efficiency throughout the organisation, and ultimately to greater profitability.
What Is An Example of Digital Twin Technology In Retail?
Especially right now, as retail looks to continue growing, carrying out customer modeling, product rollouts, and more without disrupting the business can be tremendously beneficial. Digital twin technology can help retailers to overcome some fairly unique challenges within their industry.
One major example is the challenge of providing a personalised customer experience. Shoppers continue to gravitate toward brands that “get them,” not only by delivering the products they want and need but also by representing the values and culture they care about.
No one needs to tell us that this is far easier said than done. It is difficult and risky, as top-down changes can very easily come off as inauthentic to consumers and end customers. But what if there was a way to field-test new products and branding initiatives long before they ever hit the shelf? This, I believe, is one of the many interesting use cases for digital twin in the retail space.
By creating a virtual model of each customer, retailers can gain real-world insights into their customers’ preferences and behaviour and even generate responses to proposed changes. If the virtual ecosystem doesn’t give realistic enough responses, that virtual environment can very easily be shared with both internal stakeholders and even panels and focus groups, and in much more realistic and immersive fashion than any PowerPoint deck or prototype can convey. This data can then be used to provide tailored product recommendations, promotions, and customer service.
Mapping ROI To Reality
In our talk, Joe made a great point about labour optimisation as a key way that distributors, in particular, can achieve ROI right now with digital twin. If you know how your operation runs and pair that with knowledge of your process flows, equipment run-times and labour rates, you can simulate what it’s going to be like tomorrow – or even later today – and know what your labour demands will be at that time.
This allows you to make changes now to address problems that haven’t occurred yet. Say you leverage your model to look ahead a couple of hours to the end of the current shift, and you see that one of your picking areas will be 15% behind. Armed with that knowledge, you can now proactively adjust your labour in the real world to accommodate that.
“That’s what we’ve been looking for in distribution forever,” he says. “Now we can simulate those hundred thousand web orders that are going to come down at 4 o’clock, because people like to drop those at the end of the shift.”
Are We There Yet?
Like many high-profile technologies (blockchain comes to mind as one notable example), there’s a looming question within the digital twin discussion: “Are we there yet?”
My answer to that question would be a fairly qualified yes. Ram mentioned during our talk that NASA has been using digital twin technology since they invented the concept back in the 1960s. Joe noted that in 2018, Boeing CEO Dennis Muilenberg said we were living in the “Digital Twin Era” of aviation.
One company that’s already pushing the boundaries of digital twin from a retail perspective, of course, is Amazon. As Joe mentioned during our talk, Amazon has not only twinned its own customer base, they even rolled out its “Digital Twin Maker” functionality last April, in a bid to “make it faster and easier to create digital twins of real-world systems and apply them to monitor and optimise industrial operations.”
For the rest of us, however, digital twin capabilities are finally coming into the realm of possibility. One major reason for this is the now-widespread availability of real-time data which, paired with the rapidly-maturing capabilities of deep learning systems and neural networks, is ushering in an era in which building a useful, intelligent digital model of one’s supply chain is truly feasible.
What’s in store for the future?
My predictions revolve heavily around network optimisation, predicting future outcomes and many other fairly futuristic capabilities (you’ll have to watch the full video to hear them all). While we’re not there yet – and we’re unlikely to get there anytime terribly soon – we are certainly well on our way. And as far as I’m concerned, that’s always been the most exciting place to be.
Photo: TruckPR / Flickr / CC BY-NC-ND 2.0