How Digital Twins can accelerate innovation in the CPG industry

Touchpoint 1. How can you address fragmented manufacturing operations by improving the pace and ROI of your innovation?

Digital twins facilitate continuous improvement by creating a virtual model of proposed innovations. This allows CPG companies to simulate, monitor, and optimize operations in real time without interrupting actual production. By testing new workflows, production strategies, or machinery configurations in a virtual environment, companies can identify inefficiencies, predict potential issues, and make data-driven improvements without the risk and cost of trial and error in the physical world. This accelerates innovation, reduces downtime, and helps companies adapt to market changes or customer demands more efficiently, all while minimizing operational expenses.

Touchpoint 2. How can you continuously improve operating processes?

Fragmented production capabilities limit the scope for cost-reduction initiatives and enhance productivity. Digital twins can help to continuously improve operating processes by using manufacturing data to model planned changes. This will empower you to deliver innovation at higher velocity – and lower cost and risk – to unlock value in manufacturing facilities.

“The best way to predict the future is to invent it”. The phrase is attributed to Abraham Lincoln but it is as true today as it was in the 19th century. However, innovation in the CPG industry is at an all-time low because many companies are finding that their efforts to optimize their operations and implement change in the real world are costly and ineffective. Many CPG leaders are therefore turning to Digital Twins to model existing and future processes and products in a virtual environment: this lowers the costs and risks of innovation – and increases the chances of success.

Innovation is popularly understood to be the creation of the new – when asked to think of our favorite innovation, most of us would point to a breakthrough development like the internet or the mobile phone. But McKinsey’s definition of the term casts it in a different light: `In a business context, innovation is the ability to conceive, develop, deliver, and scale new products, services, processes, and business models for customers (!)’ . So, innovation embraces the attempt to optimize ‘business as usual’ as well as the search for the next new thing.

The same McKinsey report found that over 80 percent of executives say innovation is among their top three priorities, yet less than 10 percent are satisfied with their organizations’ innovation performance. In the specific context of Consumer-Packaged Goods (CPG), the Boston Consulting Group reports that an elite 20% of CPG companies ‘stand out for their ability to leverage innovation to accelerate growth, enhance operations, and sustain a competitive edge… Such outperformers demonstrate that a robust innovation function is both strategically and commercially beneficial (!)’. 

Yet Mintel also recently reported that global CPG innovation reached an all-time low in 2024 (!). 
Major CPG brands are coming under pressure from private and white label competitors, with Mintel noting that ‘the emergence of AI and eCommerce have lowered the barriers to entry for smaller, direct-to-consumer brands, who can build their brand equity and sales online.’ It concludes that, ‘Innovation will be a central pillar for bigger CPG brands to survive and thrive… as we enter the second half of the 2020s.’

So, in an increasingly competitive environment, CPG companies are under pressure to find cost-effective ways to innovate – to optimize processes that will increase productivity and drive down costs, and to bring to market new products and services that will fuel growth.

Introducing the Digital Twin

What if there was a way of simulating real-world processes or systems, both those currently in place and those planned in the future? What if you could apply real-world data to those models to understand where problems were occurring? Or model future products or processes to understand how they interact with real-world systems? All without putting at risk the existing production facilities on which you depend.

That is precisely the role of a digital twin. These are dynamic models of an existing or planned physical object or system. They centralize data in one place to eliminate data silos and ensure that all historical and real-time data sit together. Digital twins offer a road to continuous optimization across lifecycles, including better collaboration, informed decision-making, reduced development time, accelerated risk assessment, and predictive maintenance capabilities.

Digital twins are deployed alongside other technologies –including IoT, extended reality, data analysis, and machine learning – that are being adopted in the CPG industry. The digital models are fed with real data acquired by sensors within the real process and processed using artificial intelligence (AI). Via the digital twins, users can simulate behaviors and predict how objects and processes will perform. By 2025, 80% of industry ecosystem players will leverage their own product, asset and process digital twins to share data and insight with other participants, according to IDC.

Accelerating innovation
Digital Twins can enhance innovation in CPG companies in several different ways. CPG leaders can use digital twins to evaluate the impact of various decisions and strategies, enabling more agile and effective responses to changing market conditions. And Gartner recently claimed that a Digital Twin of a Customer (DToC) – “a dynamic virtual mirror representation of a customer that organizations can use to simulate, emulate and anticipate behavior” – is beginning to make its journey along the Hype Cycle (!).  However, the two major use cases for digital twins among CPG companies are to simulate product development and to optimize manufacturing and supply chain processes.

Simulating product development: 
Digital twins allow CPG companies to create virtual replicas of their products, enabling them to test and experiment with different designs, materials, and functionalities without incurring physical costs or delays. They can also create digital twins of new product designs to simulate and test their performance under various conditions before implementing them in the physical world. Using digital twins to model the impact of different ingredients or materials on product quality, shelf life, and consumer satisfaction can also accelerate the R&D process. Overall, digital twins can reduce the time and cost associated with physical prototyping – and de-risk the entire innovation process.

In its ‘Digital twins: The key to smart product development (!)’  report, McKinsey found digital twins significantly optimize product development performance. In some cases, development time has been cut by 20 to 50 percent whilst also reducing costs. Other companies have been able to reduce the number of expensive preproduction prototypes from two or three to just one.

In addition, carrying out testing, verification, and customer-acceptance work in a virtual environment can result in products with 25 percent fewer quality issues when they enter production. Products created via digital twins may also stand a better chance of succeeding in the real world: according to McKinsey, “one company reported 3 to 5 percent higher sales of digital-twin-based products, thanks to better features, higher quality, and improved customer satisfaction.”

Optimizing manufacturing and supply chain processes:
Digital twins of manufacturing facilities can be used to identify bottlenecks, simulate different production scenarios, and optimize workflows, leading to increased efficiency and reduced costs. In the supply chain, digital twins can help CPG companies identify potential disruptions, optimize inventory levels, and improve overall supply chain efficiency. They can also be used to collaborate with partners and suppliers, enabling more efficient and effective product development and manufacturing processes.

A different McKinsey report (!) found that a factory digital twin was recently used to optimize a production schedule, resulting in monthly savings of between 5 and 7 percent by compressing overtime requirements at an assembly plant. A different factory digital twin was also able to optimize the scheduling of thousands of potential product combinations across four parallel production lines by identifying ideal batch sizes and production sequences.  

Challenges of adopting digital twins
Creating an end-to-end digital twin platform is not straightforward. A digital twin is not a single technology: it requires visualization, workflow and APIs to share data from multiple sources to provide a machine-learning framework for real-time data ingestion. However, implementing a digital twin is not purely a technology problem as, above all, it requires accurate data that is cleaned and must have a good semantic ontology (this is a structured framework that formalizes the meaning of concepts, categories and their relationships.) As a result, it is crucial to involve domain experts that can not only understand the data but also contextualize it.

In essence, a digital twin is only as good as the data on which it relies. In a legacy manufacturing environment, there may be gaps in your data; but, even without these, aggregating data from multiple systems and ensuring the consistency necessary to enable real-time decision making is challenging.

This is where a partner like Orange Business – whose experience in digital twins includes condition monitoring, predictive maintenance, automated quality control and the management of fuel, energy and water usage – can make a difference. We can guide you through the creation of your digital twin and ensure that you build simulations that best match your business requirements.

Generally, It is best to start small and scale up. Create a minimum viable product (MVP) of a digital twin that focuses on a problematic sub-process which, if resolved, will have a significant impact on your operations. This will allow you to identify best practices, avoid mistakes and acquire the insights that will inform future initiatives.

Conclusion
According to Forbes (!), high CPG innovation failure rates – as much as 85% according to Nielsen data – and the huge costs associated with this ‘are not only tolerated – but expected.’ This is simply unsustainable – responsible CPG leaders must expect more from their innovation investments. By adopting digital twin technology, CPG companies can significantly improve their agility and capacity for innovation, driving better business outcomes and staying competitive in a rapidly evolving market.