Balance your logistics strategy against customer service? Predict the impact of a natural disaster on your supply chain? Model an increase of demand through your distribution and sourcing? Digital twins can help with better predictions.
Product Digital Twins are emerging everywhere, but a Process Digital Twin is the next step. And it will be huge for Supply Chain.
Digital twins are a representation of something that exists in the real world, but then as a digital object.
The concept of using a representation as a means of research and simulation is not new. Physical models, prototypes and even digital simulations have been used for long time in product and process.
However, today’s technological evolutions allow going way beyond the product engineering use. Models can include many more parameters and can be applied to mirror and represent unique individual physical objects or even business processes. This evolution has allowed the emergence of the unique ‘digital twin’.
What Can a Digital Twin Mean for Supply Chain?
Digital twins typically combine a history of the past, a presentation of the current state and allow to predict the future. The digital twin can serve as the ultimate simulation tool or a great ‘what-if’ scenario analyzer.
Today there’s an abundance of source data, data storage and advanced computing power. Leveraging the Internet of Things (IoT), connectivity, Data Lakes, Machine Learning (ML) and modern visualization techniques, digital twins can now represent complex multi-tier global supply chain complexity.
Any supply chain digital twin should be based on a model that is the outcome of a solid Planning process. So starting with a well-coordinated Sales & Operations Planning (S&OP) initiative. The resulting Demand and Supply Chain Planning process then specifies how demand and supply balance out, including Demand Planning, Distribution Planning, Production Planning and Procurement Planning. These drive the expected parameters on revenue, distribution, sourcing, cost and profit.
And then the digital twin comes in. It allows to capture the historical data and parameters of the supply chain. How are we executing against plan ? It also represents the current status of the Supply Chain. It is a mirror of your actual processes allowing it to serve as a Dashboard. Driving action when it highlights a hiccup in the supply chain. Eg. it can highlight a discrepancy in actual demand versus forecasted demand and visually represent that for rapid action.
And – finally – the Digital Twin can also be used for simulation. Use it to continuously fine-tune your supply chain. Eg. assess the impact of a natural disaster in a certain country that wipes out a factory or a supplier. Eg. model a logistics strategy against customer service impact. Eg. predict the impact of promotions on demand.
The key will be capturing the data that impact the supply chain. These can exist in multiple sources – ERP, CRM, Supply Chain planning, Warehousing, sensors, and even unstructured information. The ability of the model to provide accurate simulation depends on the correct usage of data. That data is typically captured in a data lake. It then will be used to feed the entire supply chain model. Gartner quotes “The quality of answers and decisions generated by the software depends heavily on the quality of the data model, or how well the model represents reality at the point in time at which the answers are generated”. But, when well executed, a well thought through supply chain digital twin can drive improved business planning, improved procurement planning, improved network planning or faster time to market.
Today’s new technologies like IoT, connectivity, Data Lakes, Machine Learning and modern visualization techniques give digital modeling a boost and drive the emergence of Digital Twins for Supply Chain processes. A supply chain digital twin should always go hand in hand and build on a solid S&OP and Demand and Supply Chain Planning process. With that input a digital twin allows to understand and visualize your global supply chain processes. It enables to identify current hiccups and allows to simulate and predict for improved performance.