The COVID-19 pandemic took its toll on many world’s leading industries, and fortunately, some were more equipped to deal with its consequences:
The pharmaceutical industry was one of the most affected by the adverse effects of the pandemic. David Schleifstein, Director of Business Intelligence in Pharmascience Inc., provided us with an insight into the obstacles the pharmaceutical industry had to overcome during the pandemic, such as forecasting difficulties, and how data science and AI were quickly and efficiently used to overcome the recent challenges.
Forecasting demand drives customer satisfaction
According to a recent article by Forbes, AI and machine learning are now being used to monitor and predict epidemic outbreaks before and after they occurred. In Datavore 2021, Pharmascience’s Director of Business Intelligence David Schleifstein addressed why the application of AI in the pharmaceutical industry was inevitable:
Firstly, given the pharmaceutical industry’s highly competitive nature, accurately forecasting the demand can serve as a significant advantage to the companies. Overforecasting can lead to the wasteful overuse of the supply chain, while under forecasting can lead to not meeting consumer demand, ultimately resulting in poor customer satisfaction.
Secondly, in usual cases, traditional forecasts work well when there are no disruptions, and all operations are going smoothly, but COVID-19 circumstances proved to be anything but ordinary. Increased border control and port restrictions made predicting the arrival of material incredibly difficult. Not only that, COVID-19 hot spots made it impossible for some workers to get to work, consequently affecting the capacity of the suppliers.
Lastly and most obviously, there was a notable change in customer behavior. A universal state of panic spread everywhere, and just as we saw with toilet paper, people hoarding large amounts of medicines caused a significant strain on manufacturers. When this occurred to one manufacturer, they automatically went on backorder, so customers shifted to another supplier, who eventually had to go on backorder. Suddenly, the entire market was out of stock, and consumers were left alone without getting what they needed.
DS Skills for All: Data Integration and Predictive Modelling
“Our company needed to respond to the changes in business needs and create a process where we can use data to forecast customer demand through indicators powered by AI,” said Schleifstein.
In this context, the Pharmascience employees must be upskilled and reskilled in data science and AI to help the companies to stay competitive.
Schleifstein shared two must-have data science skills for the ambitious pharmaceutical workers:
1) Integrate external data sources with internal data sources to make sense and provide valuable insight that can be used to make forecasting decisions. This certainly is easier said than done, given that the data is unstructured and integrating it and making connections with your internal data is not a simple task.
2) AI can empower companies to process vast amounts of information and data that can later be used to make informed decisions. Schleifstein emphasizes the need for knowledge in AI and machine learning to make critical connections, ones often not easily spotted by the human eye, between vast amounts of data in real-time. Thus leveraging AI technology will enable demand planners to make more accurate predictions.
“With better forecast models that use multiple data sources and smart insight surrounding external events, you will have better modeling and more insights coming in,” concludes Shleifstein.
Text: [Huda Hafez]