The world is becoming more and more digital.We live in a world full of new technologies. Smartphones, online payment systems, 5G and much more. They all have an impact on our everyday lives, both business and personal. Especially in the ongoing global pandemic situation, we have been and continue to be forced to make changes and adapt quickly to these new technological innovations.
Big data is the coming trend in pharma.The pharmaceutical and life science industries are generally advanced with their applications and technologies. For example, DeepMind, a Google company, uses artificial intelligence to solve the problem of protein folding. It applies cutting-edge technologies to predict the 3D structure of a protein based solely on its genetic sequence1. However, in terms of operations, the entire industry is still at an early stage of adopting new technologies. The main reason for this is the strict regulatory environment in which pharmaceutical companies operate.
New technologies offer many advantages, especially in the pharmaceutical industry. By using digital technologies, pharmaceutical companies can significantly increase the transparency of their overall operations. With this increased level of visibility, executives can react much more quickly to change, use the visibility of key data to make decisions on business-critical issues, and thus improve overall efficiency and effectiveness in the operational environment. Companies that collect, analyse, refine and visualise all their data will be global leaders in the 21st century and survive the era of digital transformation.
According to the author, pharmaceutical companies could start using Big Data technologies to develop methods and practical experience to gain insights into the data generated from day-to-day business. This opinion is supported by recent studies. This article will take a closer look at a study by McKinsey. The study identifies four paradigm shifts in how data could help companies improve operational transparency and efficiency2:
Paradigm shift on the use of data.
1. real-time prognostic analysis.Most companies can extract machine data. However, performing advanced analytics with this data is still rare due to a lack of IT staff such as data scientists, data engineers and software developers. As a first step, data engineers should build an end-to-end data pipeline between machines and data storage. Then, data scientists apply advanced algorithms to understand the patterns of machines and devices. The final step consists of advanced analytics tools programmed by software developers and used by manufacturers to detect patterns and share information with other machines. Machine learning algorithms can be used to identify risks and warn users in a timely manner. With this approach, demand could be better predicted and risks identified earlier and proactively mitigated before machine breakdowns or deviations from specification occur3.
2. end-to-end simulation of the digital twin in production.A digital twin is a complex combination of technical data combined to create a virtual technological representation of a physical process or product4. From an operational perspective, digital twins can simulate the production cycle at the level of machines, labs, factories or an entire production network. The real-time simulation then allows companies to make further configurations or changes, such as adding machines or changing the schedule, to observe performance before implementing these adjustments in the real world. In this way, plant managers can optimise or reschedule parameters to see how it affects the production line without taking a risk.
3. process optimisation.Robotic process automation tools are on the rise to further automate and improve knowledge work and administrative processes. For example, programmed robots can streamline production management and provide real-time visibility of stock movements by automating the creation of product orders5. Quality control could use machine learning algorithms, such as image recognition, to count the number of pills in each bottle, check each pill for exact size and shape, or inspect the packaging for damage, and so on.6. According to the authors, current image recognition algorithms are accurate enough to keep up with humans.
4. intelligent storage and distribution.
Warehousing is still manual and labour-intensive today. Especially in the pharmaceutical industry, where much of the production is outsourced and moved to different countries. Digital innovations in warehousing and distribution are emerging in a big way. One of the promising technologies is RFID (Radio Frequency Identification). RFID offers, for example, the possibility to track medicines from production to delivery. Each RFID tag or marker is stuck on the product and programmed with a unique code. The unique code is entered into the companies' databases and can be matched with other unique data such as shelf life. All this information is consolidated in a single ERP system and can provide real-time information and transparency about each individual product. In addition, all data exchange can be visualised and presented to the production manager7.
Success factors and risks.Companies in the pharmaceutical industry are faced with the challenge of digitising their operations in order to cope with more complex situations and keep up with their competitors around the world. They need to develop solutions not alone, but in coordination with their partners along the entire value chain, as technology is changing too fast to meet daily technical standards. To meet this challenge, companies should be open enough to change and must overcome the resistance of employees or organisations that have become too accustomed to the processes and tools they have been working with for a long time8.
Collaboration and cybersecurity are key.There are success factors that are important for the change process. Companies need to start with a clear understanding of the data ecosystem and the technologies involved. They also need to have the right resources, whether internal experts or external resources, to support sustainable implementation. Working with external partners such as research institutes, universities or consultancies is key to developing lean operational end-to-end solutions, from strategic concepts to detailed implementation and roll-out. At the same time, companies should also consider the risks and concerns. The biggest concern for pharmaceutical companies is cybersecurity. Pharma executives are concerned about the risk of hackers gaining control of physical assets and hardware. In addition, ensuring data security, especially for sensitive data such as pricing, machine performance or contracts is necessary and should be considered first and foremost before starting an implementation.
Four steps for leaders in pharma towards digitalisation.
Pharmaceutical companies should start digitisation as early as possible. The earlier companies decide to make the change, the greater their lead over competitors.
There are four steps that leaders could start with:
- First, the core operational team should engage with data experts to explore the technology landscape and develop a deep understanding of the opportunities for using data to digitise the entire operational ecosystem
- Once the technology is in place, companies should define a data strategy for the next three to five years. Assess available technology and define business cases that could save operational costs
- First, concepts for specific business cases have to be tested. Then the results are to be evaluated and a concept created
- Successful projects should be expanded and rolled out throughout the company
Well-prepared companies will survive.
The pharmaceutical industry is at the beginning of digitalisation. Digital working in combination with Big Data technologies is the innovation trend for the next decades. Only well-prepared companies will be able to be more agile and proactive, more cost-efficient and, above all, with higher operational efficiency, transparency and maturity.
- A.Senior (2020): AlphaFold: Using AI for scientific discovery, available under: https://deepmind.com/blog/article/AlphaFold-Using-AI-for-scientific-discovery, retrieval date 06.08.21
- McKinsey (2018): How data is changing the pharma operation world, available under: https://www.mckinsey.com/business-functions/operations/our-insights/how-data-is-changing-the-pharma-operations-world, retrieval date 06.08.21
- Exelpros (2020): Predictive analytics in the pharmaceutical industry: Key Use Cases https://xcelpros.com/predictive-analytics-in-the-pharmaceutical-industry-key-use-cases/, retrieval date 06.08.21
- R.King (2019): An introduction to digital twins, https://www.rowse.co.uk/blog/post/an-introduction-to-digital-twins, retrieval date 06.08.21
- UiPath : Automation solutions for manufacturing, https://www.uipath.com/solutions/industry/manufacturing-automation, retrieval date 06.08.21
- Kantify: AI Applications in pharma and biotech, https://www.kantify.com/insights/ai-applications-in-pharma-and-biotech, retrieval date 06.08.21
- PwC, Digitization in pharma, https://www.strategyand.pwc.com/gx/en/insights/2016/digitization-in-pharma/digitization-in-pharma.pdf, retrieval date 06.08.21
- RARain: Evolving Pharma and Healthcare, how RFID is setting the standard in safety, compliance and logistics, https://rfrain.itsupportme.by/blog-posts/evolving-pharma-and-healthcare-how-rfid-is-setting-the-standard-in-safety-compliance-and-logistics/, retrieval date 06.08.21.