The cloning of prehistoric reptiles did not serve well to mankind in the silver screen. However, off reel-life, the real-life application of digital clones of products, services, and a whole process framework is surely opening new frontiers of innovation in the business space. With the National Research Foundation unveiling the digital twin of the urban city Singapore, named as the Virtual Singapore, there is no doubting the fact that digital twin technology is the next big thing to count on in several aspects of an industry.
The concept of digital twin technology has its origin in as early as 1777 with Buffon’s needle problem. Simulation, which forms the basis of the digital twin technology, was leveraged by Comte de Buffon to conduct an experiment that involved repeatedly tossing a needle onto a ruled sheet of paper. This experiment was aimed at determining the probability of the needle crossing one of the lines.
Digital twin technology made its first major inroad into the real-world in the 1970s during the Apollo 13 venture undertaken by NASA. Engineers built a digital twin model of Apollo 13 on earth which aided engineers to test possible solutions from ground level. Despite being a doomed venture, the Apollo 13 fiasco heralded the gradual proliferation of digital twin technology in the business world.
Twinning in Product Manufacturing
The prospects of avoiding production downtime and cost-savings are driving manufacturers towards adopting digital twin technology to improve and streamline their product development and quality control processes. Products such as aircraft engines and excavation automobiles, interact differently in a diverse set of environments. For instance, factors such as stress analysis and fuel efficiency of excavation automobiles, vary across rocky topography and plain terrains. This makes predicting the chances of wear and tear over the life cycles of different types of machinery extremely dubious. A digital clone can ease the uncertainty of the workings of these products, by simulating possible interaction with varying topographies, among other scenarios. Gaining product intelligence to simulate scenarios would require collecting data from the sensors fitted into the physical product. The data would then be fed into a digital replica that would offer actionable insights into possibilities of wear and tear and specific kinds of stress when the product is deployed. Machine learning algorithms applied to production data can aid to detect correlations and form predictions on the remaining useful life cycle of the products. These insights are instrumental in boosting functions such as performance analysis and predictive maintenance. An improvement in these functions has historically saved manufacturers both time and money.
Today, a majority of OEMs are including a digital twin as a part of their service package. As a matter of fact, manufacturing enterprises are finding implementations of scores of smaller digital twins useful, which also calls for extremely efficient management.
GE’s Predix platform is a good example of a dexterous use of digital twin technology in the manufacturing sector. Meanwhile, to mobilize this technology, software companies such as SAP and Microsoft have developed IoT and AI-enabled platforms.
Relevance in CPG Product Manufacturing
It is not a new fact that consumers occupy the top tier of the CPG industry value chain and their dynamic expectations continue to disrupt this industry. A desperate bid to stay responsive to these expectations often wreaks havoc in the CPG product manufacturing process. Customizing products to appease consumer expectations is an extensive process that is time and cost-intensive. However, dynamic consumer behavior is just the tip of the iceberg that continues to be a strong obstacle for the smooth functioning of the CPG product manufacturing process.
How to Leverage Digital Twin Technology to Ease Dynamisms in the CPG Product Manufacturing Lifecycle
Tracking customer behavior
Digital twin technology can be used to garner real-time insights on dynamic behavioral traits. For instance, transactional information of customers such as purchase frequency to non-transactional data such as work location, home address, etc., can be leveraged to create digital clones of individuals. This can help to shed light on future purchase patterns and/or activities of customers. Proactive demand forecast alerts can help manufacturers to streamline production processes.
Redesigning a product
The main mantra to stay relevant in the hyper-competitive CPG industry is to be flexible to changing market demands. Consumables, in particular, tend to trend with customer expectations leaning towards variations in flavor, packaging, color, and shape. Normally, it takes a lot of manual trials and customer surveys, to conclude the right solutions to cater to such demands. Digital twin technology can help replicate the entire production process and run trials in a virtual atmosphere with changing parameters to narrow in on optimal product/plant settings. From a quality testing perspective, process-based machine learning can be used to trigger predictive alerts on any possibilities of deviations from the quality gold standards and their root causes.
Supply chain management
It is safe to say that in today’s world, the e-commerce site is the most-preferred transaction platform in the CPG industry. A report published by consumer goods data-collection firm Information Resources Inc in early 2019 claimed over a 30% rise in the online sales of CPG products in the US. With a multitude of e-commerce outfits at consumers’ disposal, expectations on shopping experience have risen over time, in the context of product quality, availability, and timely delivery. While these demands have mounted pressure on CPG supply chains, it has reinstated the importance of stepping up the efficacy of operations. An optimized digital twin solution is critical, in this regard, as it replicates the end-to-end supply value chain and offers keen insights on possible product defects, plant inefficiencies, among others.
A seamless co-execution of data mining techniques and Internet of Things (IoT) sensors allows transmission and ingestion of data from a physical supply chain system to its digital counterpart. Digital twin technology is the obvious next step from a prescriptive analytics model in supply chain management, as it enables feeding data into the model in real-time. For instance, a digital supply chain model is automatically fed with information on the purchase order, associated transactions, and more, when a customer makes a purchase. , Data related to a possible production downtime or a delay of the cargo ship can also be fed to the digital twin of the supply chain, by monitoring machinery and shipping vehicles. Generally, IoT devices are used to detect this information and enable the digital twin to determine contingent measures to avoid possible disruptions. DHL is known to leverage a digital twin warehouse which is programmed to ingest real-time data from its physical counterpart. The data is then conveyed to the concerned team who analyses it to identify potential measures to optimize storage solutions.
The application of digital twin technology is still in a nascent stage with enterprises taking slow but steady paces towards adopting it in the CPG industry. Hence the question that CPG businesses should be asking themselves is not about whether they should adopt digital twin technology, rather how they could go about integrating digital twin technology with their existing ecosystem. Of course, building separate digital twins for different segments in the product manufacturing framework will not only entail costs but also trigger operation management difficulties. So, will it still be a prudent business decision to leverage digital twin technology in CPG product manufacturing? If yes, then what should be the best practices to derive the maximum value from investment in this technology? Stay tuned for the sequel to this article that will answer these imposing questions.