THANK YOU FOR SUBSCRIBING
Hacking Business-How Big Data and Artificial Intelligence are Changing Business and the Pharmaceutical Industry
Philipp Diesinger, Chief Data Officer, Boehringer Ingelheim International GmbH Data Science
In the past, the application of this scientific modelling approach has mostly been constrained to the fields of science and academic research. Models are developed from observations (“data”) which had to be collected in specifically set up and often times complex experiments.
This has fundamentally changed the ability to collect and store vast amounts of data in a cheap and reliable way has been improving dramatically over the past decade. Currently, the human race is collecting data at a rate of about 2.5 Exabytes (10^18 bytes) per day. The total data of humankind is expected to grow ten-fold within a few years until 2020. New technologies, infrastructure, and intelligent solutions have been improving exponentially and thus opened up the world of business for the first time on a large scale for scientific data-driven and model-driven decision making. This new approach to business is called “data science”.
On top of things, information is flowing faster than ever before to an extent that makes it impossible for humans to keep up. Many business decisions, such as financial transactions or online advertising, need to be made in real-time (ms) today in order to gain a competitive advantage and not miss out on critical business opportunities.
All of the sudden, companies are facing the challenge of having to find new ways to leverage large amounts of internal and external data for their business. Luckily, in order to generate business value from data, one can utilize a vast library of already existing analytical approaches which have been established and refined over decades (sometimes even centuries) in academia.
These methods typically include the fields of statistics and probability theory in mathematics and computer science as well as first principle or stochastic models in theoretical physics.
Check Out: Life Sciences Review Magazine
However, the large amounts of data available today combined with improvements in processing power and GPU technology also make relatively new approaches feasible for large scale applications for the first time now. These new technologies include methods like causal inference to extract insight from data, machine learning approaches like decision trees or regression models as well as deep learning techniques based on the complexity of graphs. Last but not least, the new ability to process and analyze unstructured data such as text, audio or image data also opened up room for novel analytical approaches such as natural language processing or speech recognition. All of these opportunities are expected to drive disruptive changes in business environments around the globe and across industries over the next decade.
Organizations in firm control of their data will realize a big competitive advantage in this new era of technology
In a traditional field like the pharmaceutical industry with its highly regulated market on one side and significant cost pressure on health care systems on the other side, new technologies open interesting and promising business opportunities. For instance, first studies show that artificial intelligence may soon play an important role in small molecule computer-aided drug discovery. Not only a large variety of pills but also bone structures and skin tissue can already be 3D printed and thus manufactured on site and more importantly adapted to the specific needs of individual patients. 3D printing is also expected to have a large impact in the field of medical equipment which is currently a significant cost driver in the healthcare sector with many items used during surgical procedures being restricted to one-time use only. Furthermore, the way we engage with health care practitioners is changing. Decentralized online networks of doctors are forming around the globe and care is being more and more personalized using large amounts of genetic and other omics data. Blockchain technology may become a key driver for change in the healthcare insurance business. New data sources, such as electronic medical records open up new approaches in clinical research in order to better understand diseases and the treatment of patients and intelligent recommendation systems may provide health care practitioners with patient-specific treatment suggestions in the future.
However, new technologies are not limited to disruptive changes in business models but can also improve decision making within large organizations dramatically. For instance, AI systems in combination with IoT, smart sensors and predictive maintenance can have a significant impact on the production of pharmaceutical drugs where quality standards are high leading to dramatic improvements of efficiency, reduction of downtime and utilization of cost savings. Furthermore, AI recommendation systems have shown to significantly improve financial decision making such as budget allocation, portfolio management or optimal tendering strategies in combination with game theoretical approaches.
Current management is facing the challenge of having to deal with these novel approaches, integrating them into business decision making as well as adapting business models to identify and leverage new opportunities. The field of data science is developing so rapidly that companies who traditionally rely on the next generation of management to deal with these changes are expected to be at a significant competitive disadvantage. In contrast, organizations in firm control of their data will realize a big competitive advantage in this new environment.