As edge computing continues to rapidly grow, retail operators will have greater access to AI-driven insights and digital replication. The incorporation of digital twins into operational processes is accelerating capabilities for real-time response and automated decision-making.
Gartner defines “digital twin” as “a digital representation of a real-world entity or system.” Digital twins compile data from the physical world and generate multiple simulations and predictions.
With the potential to be more than simply a snapshot or model, digital twins’ multi-simulation nature exponentially increases the speed at which predictive and prescriptive analytics can be leveraged to optimize operations. The increasingly instantaneous feedback creates numerous opportunities for the advancement of asset protection and operational efficiency.
DIGITAL TWINS FOR THE SUPPLY CHAIN
Numerous applications of the digital twin concept have been developed and applied in various industries. For supply chains, digital twins are used to identify disruptions that cause overstocking or understocking of inventory. Applied to a supply chain network, digital twins represent the flow of goods by analyzing data from a long list of considerations:
- Location of warehouses
- Historical demand coming to warehouses at every node
- Amount of inventory on-site at the warehouses
- Where the inventory is coming from
- Where the inventory is being shipped to
- Mode of transport—air, land, sea
- Source of transport—owned, third-party logistics
- Cost of transport
- Lead time for production, transport, and delivery
- Replenishment cycle and real lead time to every node in the supply chain
- Product loss
- Manufacturing country of origin
- Component suppliers
- Demand forecast, forecasting error, and standard deviation
In this demand-driven supply network, the complexity of the flow of goods, from raw materials to assembly to production to delivery, is subject to disruptions at every stage. What if production is stopped because of outbreaks or malware? What if demand in Europe increases to 50% and demand in the U.S. rises by 20% at the same time? What if a raw material supplier is late with their delivery or discovers a major quality issue? All of these inputs and interruptions affect the decision-making process as costs, risks, and benefits are weighed.
The data generated while answering these various questions comes from multiple sources and systems. As companies integrate their systems, the data begins to form a snapshot of the supply chain network. That snapshot is a digital representation—or digital twin—of the physical activities taking place within the supply chain.
However, it is a “live” system—one could take snapshots in the future based on specific simulation of parameters. As the source and speed of the data improves, the digital twin becomes increasingly identical to its physical counterpart. Therefore, you could ask “what if?” questions:
- What if my supplier gets malware and is “down” for two weeks?
- What will be my time to survive?
- What will be my time to recover?
Answering these questions will help minimize the risk of any disruption throwing a wrench into your demand-driven supply network.
REAL-TIME DIGITAL REPLICATION FOR ASSET PROTECTION
Moore’s law is constantly pushing the boundaries of hardware. With protocols like LoRaWAN, 5G, and REDCap improving at a rapid pace, and the growing capabilities of low-power devices, edge computing is going to become much stronger. More computing power at the edge enables AI to drive real-time data to accelerate the accuracy of digital twins.
When IoT sensors collect data from a connected device, the actionable insights can be used to update a digital twin of the device’s state in real time. With enough sensors, not only can a digital twin of the device be generated but also a digital twin of the asset that the device is monitoring. Imagine that an asset is being moved from one spot to another, and the feedback on its location, temperature, humidity, pressure, CO2, oxygen, and movement is instantaneously syncing with the digital twin. The ability to intervene in real time based on in-the-moment information is possible.
With real-time replication of the physical status of assets—walk-in freezers, refrigerators, promotional coolers, shelving, carts, and other equipment—and automated corrective actions, it’s easier to avoid wasting electricity, time, and product when anomalies arise. Digital twins promote positive outcomes for asset protection programs through reduced waste and loss and, most importantly, saved lives.
UPGRADING OPERATIONS
Supply chain professionals and manufacturers have long been leveraging the capabilities of digital twins to improve performance. Operators have an opportunity to leverage those same capabilities to enhance productivity and increase profitability. Digital twins are used to optimize maintenance schedules, enrich prescriptive workflows, inform advanced resource planning, and improve loss-prevention outcomes.
Real-time data and 360-degree visibility reduce false positives and refine the accuracy of guidance to employees. With the right application, more data does not mean more reports. Operators are already overloaded with too many reports, inaccurate alerts, and never-ending notifications. The promise of digital replication of the physical world is increased accuracy and guided interventions that lead to better outcomes. When leveraging the right system, more data means more intelligence, making operations more efficient and asset protection more systematic.
Guy Yehiav is President of SmartSense.