If the value of digital transformation was not clear before, then this pandemic has sounded a wake-up call for its importance in advancing innovation and responding to surprises. Life science companies that are already fully digitized have a clear advantage—to keep their research moving forward during the pandemic, and for some companies, quickly shift to developing new treatments and vaccines for Covid-19. But the advantages of digital transformation are greater and longer lasting-they will pave the way for new technologies and new capabilities, and fundamentally change the way science is carried out.
In many industries, an increasingly compelling benefit of digital transformation is the ability to build a digital twin—that is, a complete Computer simulation A copy of the actual structure, instrument, or process.Organizations in industries other than biopharmaceuticals have been using this concept for many years, from NASA arrive give arrive BoeingThe aviation industry uses digital twin models to model engine operation and predict its function and service needs over time. This is an accurate, cheap (and of course safer) method of predicting future events.
For any industry, a digital twin is a very effective way to understand your processes and to happen ahead of time outside of normal operations. The amazing advances in the Internet of Things (IoT) and artificial intelligence (AI) are driving digital twins, which are the driving forces of industrial automation and smart factory concepts. 2020, global The digital twin market is valued US$3.1 billion is expected to grow to more than US$48 billion by 2026. Perhaps not surprisingly, the pandemic has stimulated demand for digital twins in the healthcare and pharmaceutical industries.
For the biological process industry, digital twins also have great potential. Over the years, we have become more and more automated, but the concept of creating a digital copy of a process in a laboratory—maybe one day, an entire laboratory or manufacturing plant—is very attractive. Since the timeline from discovery to market is still up to 15 years and costing billions of dollars, predicting how the drug will work, the safety issues that may be involved, and how to scale up production strategies can significantly improve processes and reduce market entry and exposure The time required by the patient.
For several years, the industry has been working towards a digital twin. But there are important differences between jet engines and drug development, especially biologics-biological processes are extremely complex, and we have not fully understood the countless nuances of human cells. It is difficult to model and simulate what you don’t fully understand-but we are getting closer and closer every day.
One method we can achieve is to collect and organize data under different conditions. Disrupting the cellular environment in many ways, or changing the conditions in the process, instrumentation, and workflow, and measuring the results, can build massive amounts of data, allowing you to start creating accurate models.
Once you have a lot of data, it is difficult to understand it, especially if your company is not fully digital. The reality is that about 50% of companies still mainly use Excel or even paper to record data—even if they are partially digitized, data is often stored in data islands, lakes, or warehouses that are not fully integrated.
Therefore, we need to move from old data storage and management systems to systems that manage data in an easy-to-understand manner and provide integrated and contextualized systems. Adopting this type of system—Biopharmaceutical Lifecycle Management System, or BPLM—not only enriches data collection and analysis, but also makes it possible to create digital twins in biopharmaceuticals.
The key is that BPLM creates a comprehensive data backbone throughout the development lifecycle-data is contextualized because it is collected at the point of generation, across multiple stages and processes. Then you can integrate data from different instruments, workflows, and departments in an efficient and structured way. Importantly, an important source of wasted time and money in R&D—rework due to data loss—has been significantly reduced. The end result is that the data is easier to understand and gain insights from it.
Similarly, an important advantage of the BPLM system is that it makes digital twin generation possible: the captured data can “train” the digital twin and make them valuable predictive and even troubleshooting tools. As we develop more and more complex therapies and technologies, digital twins will become more valuable. For example, due to the complexity involved in many stages, mRNA vaccine development is particularly suitable for this process. Digital copies can help optimize development. An obvious example is that scientists can adjust the mRNA code on the computer when new variants appear, to create an updated version of the vaccine as a booster.
Another benefit of digital twins is that it opens the way for smaller biotech companies-if more work is done digitally and predictively, the time and cost of bringing drugs to market will be reduced. In this way, the competitive environment for promising young companies that want to compete with the giants becomes flat-and because some of these small companies have integrated into their digital strategies from the beginning, this allows them to gain in the growth process. Advantage. But for companies of any size, using digital twins can reduce costs and increase efficiency: just reducing work and rework can save millions of dollars.
In essence, BPLM is at the core of our future and biopharmaceutical 4.0. By collecting and analyzing data in a completely new way, BPLM is a “hub” that promotes cross-system integration with the Internet of Things, and commercializes artificial intelligence and advanced analytics. It takes partners to work together to fully realize this integration-first across laboratory instruments, then laboratories and facilities, and then a day across disciplines.
It is almost certain that this digital direction will accelerate rapidly in the next ten years, whether in the industrial field or in daily life. We already enjoy it: smartphone apps constantly make predictions, and even simple cars check whether their drivers try to change lanes without a signal or get too close to the vehicle in front. Those are twins, constantly and silently predicting and adjusting. For biological processes, we will also achieve this goal, starting small, and then moving towards the digital twin of everything. It should be noted that, by definition, simulations are not 100% accurate-but we can get closer and closer, not only considering physical effects through the laws of nature in the model, but also improving them by collecting more contextual data. The quality of the data will always predict the quality of the simulation. This is true real-time modeling-it will provide unprecedented predictive capabilities in the coming years.
